Quantitative Imaging with a Mobile Phone Microscope
Open accessAbstract
Use of optical imaging for medical and scientific applications requires accurate quantification of features such as object size, color, and brightness. High pixel density cameras available on modern mobile phones have made photography simple and convenient for consumer applications; however, the camera hardware and software that enables this simplicity can present a barrier to accurate quantification of image data. This issue is exacerbated by automated settings, proprietary image processing algorithms, rapid phone evolution, and the diversity of manufacturers. If mobile phone cameras are to live up to their potential to increase access to healthcare in low-resource settings, limitations of mobile phone-based imaging must be fully understood and addressed with procedures that minimize their effects on image quantification. Here we focus on microscopic optical imaging using a custom mobile phone microscope that is compatible with phones from multiple manufacturers. We demonstrate that quantitative microscopy with micron-scale spatial resolution can be carried out with multiple phones and that image linearity, distortion, and color can be corrected as needed. Using all versions of the iPhone and a selection of Android phones released between 2007 and 2012, we show that phones with greater than 5 MP are capable of nearly diffraction-limited resolution over a broad range of magnifications, including those relevant for single cell imaging. We find that automatic focus, exposure, and color gain standard on mobile phones can degrade image resolution and reduce accuracy of color capture if uncorrected, and we devise procedures to avoid these barriers to quantitative imaging. By accommodating the differences between mobile phone cameras and the scientific cameras, mobile phone microscopes can be reliably used to increase access to quantitative imaging for a variety of medical and scientific applications.
AI snapshot
Written from the paper’s full text.
TL;DR
A custom multi-phone CellScope microscope can achieve micron-scale quantitative imaging with mobile phones, but automatic focus, exposure, color gain, and image processing can degrade resolution, color accuracy, and feature size unless corrected. The authors characterize these effects across iPhones and Android phones from 2007–2012 and propose calibration and capture procedures to enable reproducible mobile-phone microscopy.
Key findings
- With a 10×/0.25 NA objective and 20× widefield eyepiece, phones with >5 MP—typically effective pixel spacing of 0.44–0.47 µm at the sample—capture nearly all resolution information collected by the microscope optics; Apple and Android performance was similar, and iPhone resolution improved 63% over five years and six models.
- Resolution improves with objective NA but remains worse than the Rayleigh criterion; increasing from 40×/0.65 NA to 60×/0.85 NA gave no measurable resolution gain, while moving from 10×/0.25 NA to 40×/0.65 NA improved resolution from 1.74 µm to 0.87 µm but reduced field-of-view diameter from 812 µm to 205 µm, a 16× area reduction.
- Illumination varied by <7% across a clear field, distortion remained <8%, and pixel response was nonlinear but recovered linearity after gamma 2.2 correction (R² = 0.999).
- Phone automation caused artifacts: autofocus changed measured feature size by as much as 6%; auto-exposure/gain changed apparent red-to-green transmission through pinholes from 10:1 to 1:1; sharpening produced edge overshoots/undershoots greater than 5% of normalized response.
Methods
- The CellScope mobile phone microscope used a finite conjugate transmission path with a 10×/0.25 NA objective, 20× widefield eyepiece, folding mirrors, 3D-printed ABS casing, LED flashlight with diffuser, and adjustable phone holder; reference optics used a monochrome configurable CMOS camera oversampling at ≥4× Nyquist.
- Phones tested included all iPhones from 2007–2012 (iPhone through iPhone 5) and Android phones HTC G1, HTC Wildfire S, Samsung Galaxy Ace, and LG Nexus 4.
- Resolution was measured with a 1951 USAF target using Michelson contrast >10% for vertical and horizontal bars; distortion used a 32 line-pairs/mm bar pattern; uniformity and linearity used a 540/40 nm filter and calibrated spectrometer; contrast experiments used the Almost DSLR app on an iPhone 4 to control exposure, white balance, and focus.
Limitations
- Stated: certain on-board image processing cannot be disabled, so phone-specific artifacts remain; manufacturer or software updates may require re-standardization; the specific 5 MP threshold corresponds to the particular objective and eyepiece combination used.
- Stated: the condenser NA is imperfectly defined and approximated as ∼0.4; three-bar targets have known limitations for image-quality determination.
- Evident from the text: the study is limited to phones released 2007–2012 and largely to default camera functionality; proprietary demosaicing and filter-array layouts are unknown, so a common Bayer array is assumed, and results may not generalize to newer phones or other optical configurations.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 31 cited sources checked · 0 supported · 0 flagged
- Findingown data
The iPhone 4 and a scientific camera agree well for most objectives.
“We find that the iPhone 4 and scientific camera agree well for most objectives, indicating that a 5 megapixel mobile phone with a typical ∼4 mm focal length and f/2.8 lens is capable of capturing the resolution information collected by a representative low-cost microscope system using a 20x eyepiece.”
Quote in the paperSubstantial claim - Findingreasoning
The condenser NA of the mobile phone microscope is approximately 0.4.
“Since the transmitted-light illumination on the mobile phone microscope is from a simple plastic diffuser rather than a more complicated and expensive Kohler illumination system, the condenser NA is imperfectly defined, but is ∼ 0.4 for our system based on the angle defined by the radius of the plastic diffuser and its distance from the sample.”
Quote in the paperSubstantial claim - Findingown data
Distortion remains below 8% in the image.
“We utilized a regularly spaced line grating to measure the spatial dependence of magnification and find that distortion remains below 8% in the image (Figure 3B).”
Quote in the paperNot a substantial claim - Findingown data
The total detected intensity varies by less than 7% across a clear field of view.
“We find that the total detected intensity varies by less than 7% across a clear field of view (Figure 3A), due to a combination of non-uniform illumination from the single LED, sensor sensitivity, and vignetting, among other factors [26].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [26] Smith WJ (2008) Modern Optical Engineering. 4th ed. McGraw-Hill.
Not checked in the free run.
- Findingown data
Using all versions of the iPhone and a selection of Android phones released between 2007 and 2012, phones with greater than 5 MP are capable of nearly diffraction-limited resolution over a broad range of magnifications, including those relevant for single cell imaging.
“Using all versions of the iPhone and a selection of Android phones released between 2007 and 2012, we show that phones with greater than 5 MP are capable of nearly diffraction-limited resolution over a broad range of magnifications, including those relevant for single cell imaging.”
Quote in the paperSubstantial claim
- Findingown data
Quantitative microscopy with micron-scale spatial resolution can be carried out with multiple phones.
“We demonstrate that quantitative microscopy with micron-scale spatial resolution can be carried out with multiple phones and that image linearity, distortion, and color can be corrected as needed.”
Quote in the paperSubstantial claim - Findingown data
Older mobile phones with lower pixel count sensors and less well-corrected camera lenses fail to capture detail around the nucleus of the granulocyte.
“For example, the pixelated images taken with older mobile phones having lower pixel count sensors – and also typically having less well-corrected camera lenses incorporating fewer optical elements – fail to capture detail around the nucleus of the granulocyte.”
Quote in the paperNot a substantial claim - Findingreasoning
Increasing objective numerical aperture and magnification from 40×/0.65 NA to 60×/0.85 NA does not yield a measurable increase in microscope resolution.
“Consistent with this, we find that increasing objective numerical aperture and magnification from 40×/0.65 NA to 60×/0.85 NA does not yield a measurable increase in microscope resolution, likely due both to insufficient aberration correction and to the reduced resolution improvement expected when the objective NA increases beyond the condenser NA (Figure 2A).”
Quote in the paperSubstantial claim - Findingown data
The diameter of the field of view decreases by a factor of four (from 812 μm to 205 μm) between the 10×/0.25 NA and 40×/0.65 NA objectives.
“While measured resolution improves by a factor of two (from 1.74 µm to 0.87 µm), there is a decrease in the diameter of the field of view by a factor of four (from 812 µm to 205 µm).”
Quote in the paperSubstantial claim - Findingown data
Measured resolution improves by a factor of two (from 1.74 μm to 0.87 μm) between the 10×/0.25 NA and 40×/0.65 NA objectives.
“While measured resolution improves by a factor of two (from 1.74 µm to 0.87 µm), there is a decrease in the diameter of the field of view by a factor of four (from 812 µm to 205 µm).”
Quote in the paperSubstantial claim - Findingcitation
A linear response to incident intensity can be recovered on a mobile phone by conducting a gamma transformation with an exponent of 2.2.
“Notably, a linear response to incident intensity can be recovered on a mobile phone by conducting a gamma transformation with an exponent of 2.2 [27], corresponding to the inverse of the gamma encoding algorithm for sRGB images (Figure 3C).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [27] Ebner M (2007) Color Constancy. 1st ed. West Sussex, England: John Wiley and Sons. source
Not checked in the free run.
- Findingown data
The same region of a blood smear imaged across a range of mobile phones coupled to the same mobile phone microscope shows visible differences in resolution, brightness, and color balance.
“While the various phones provide similar morphological information, there are visible differences in resolution, brightness, and color balance (Figure 1D).”
Quote in the paperSubstantial claim - Findingown data
Changes in the spectral composition of the light source are often more complex than can be corrected for by three-parameter sensitivity adjustment.
“Changes in the spectral composition of the light source are often more complex than can be corrected for by this type of three-parameter sensitivity adjustment (Figure 6D–F).”
Quote in the paperSubstantial claim - Findingcitation
The high quality storage setting of most phones fulfills the DICOM requirement for compliant imaging.
“DICOM guidelines indicate that a file size reduction of 15–20x for storage can maintain diagnostic quality [43], and the high quality storage setting of most phones fulfills this requirement for compliant imaging (Figure 7E).”
Quote in the paperNot a substantial claim - Findingown data
An example of sharpening artifact is the apparent haloing around high contrast chrome features in a USAF target captured by an HTC Wildfire S.
“An example of this sharpening artifact is the apparent haloing around high contrast chrome features in a USAF target captured by an HTC Wildfire S (Figure 6G).”
Quote in the paperNot a substantial claim - Findingown data
Automatic focus adjustment changes the effective magnification, thereby changing the measured feature size of a microscope image by as much as 6%.
“This automatic adjustment does however result in changing the effective magnification, thereby changing the measured feature size of a microscope image by as much as 6% (Figure 6B).”
Quote in the paperSubstantial claim - Findingown data
The edge profile of both mobile phones deviated from a monotonic profile on both sides of the edge, with overshoots and undershoots of intensity greater than 5% of the normalized response.
“However, the edge profile of both mobile phones deviated from a monotonic profile on both sides of the edge, with overshoots and undershoots of intensity greater than 5% of the normalized response (Figure 6H).”
Quote in the paperSubstantial claim - Findingown data
Regardless of manufacturer, modern phones with five or more megapixels, typically corresponding to effective pixel spacing between 0.44-0.47 μm at the sample plane, are able to capture nearly all of the resolution information collected by the microscope with a 10×/0.25 NA objective and 20x widefield eyepiece.
“We note that regardless of the manufacturer modern phones with five or more megapixels, typically corresponding to effective pixel spacing between 0.44–0.47µm at the sample plane, are able to capture nearly all of the resolution information collected by the microscope with a 10×/0.25 NA objective and 20x widefield eyepiece (Figure 4B).”
Quote in the paperSubstantial claim- The authors state this as an assumption: The mobile phone camera sensor uses a common Bayer Filter Array directly on the sensor.
- Findingown data
Intensity profiles at edges show phone-specific deviations from the expected theoretical monotonic profile obtained from an incoherently illuminated sample using a scientific camera and no image processing.
“Intensity profiles at edges show phone-specific deviations from the expected theoretical monotonic profile obtained from an incoherently illuminated sample using a scientific camera and no image processing.”
Quote in the paperSubstantial claim - Findingown data
The study demonstrated the difficulty of obtaining a reproducible, spectrally accurate response across multiple samples for quantitative purposes when using the default camera functionality of the phone.
“We then demonstrated the difficulty of obtaining a reproducible, spectrally accurate response across multiple samples for quantitative purposes when using the default camera functionality of the phone.”
Quote in the paperSubstantial claim- Rests on an assumption: The potential of mobile phone microscopy has been widely recognized.
- Findingown data
Image artifacts caused by automated algorithms can be minimized through a set of corrective procedures implemented as part of the mobile phone microscopy workflow.
“However, we find that image artifacts caused by automated algorithms can be minimized through a set of corrective procedures implemented as part of the mobile phone microscopy workflow.”
Quote in the paperSubstantial claim - Findingown data
Phone auto-focus changes effective magnification for samples held at different distances from the objective lens, resulting in changes to apparent feature size.
“Phone auto-focus changes effective magnification for samples held at different distances from the objective lens, resulting in changes to apparent feature size.”
Quote in the paperSubstantial claim - Findingcitation
The edge profile produced by the CMOS camera demonstrates the expected monotonically increasing edge associated with the diffuse incoherent light source used in the system.
“The edge profile produced by the CMOS camera demonstrates the expected monotonically increasing edge associated with the diffuse incoherent light source used in the system [25].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [25] Inoue S, Spring KR (1997) Video Microscopy: The Fundamentals New York, N.Y.: Plenum Press. 26–44. source
Not checked in the free run.
- Findingown data
Considering only Apple phones over time, the resolution of mobile phone microscopy with iPhones has improved 63% over the past five years and six phone models.
“Considering the performance of only Apple phones over time, the resolution of mobile phone microscopy with iPhones has improved 63% over the past five years and six phone models (Figure 4C), together with increasing pixel number and other technological improvements to the phone camera.”
Quote in the paperNot a substantial claim - Findingown data
Saturation in individual color channels due to auto-exposure and gain causes loss of color contrast in sparse, bright samples.
“Saturation in individual color channels due to auto-exposure and gain causes loss of color contrast in sparse, bright samples.”
Quote in the paperSubstantial claim - Findingnone given
The camera of a mobile phone automatically adjusts focus in traditional photo capture mode by altering the lens focal length and thus the optical magnification.
“The camera of a mobile phone automatically adjusts focus in traditional photo capture mode by altering the lens focal length (Figure 5C) and thus the optical magnification.”
Quote in the paperSubstantial claim - Findingnone given
Mobile phone cameras have a more opaque control scheme for color data that involves automatic exposure and color channel gain.
“Mobile phone cameras, however, have a more opaque control scheme for color data that involves automatic exposure and color channel gain (Figure 5C).”
Quote in the paperSubstantial claim - Findingown data
Built-in white-balance of the iPhone 4 is insufficient to overcome variable illumination conditions, resulting in a shift of the apparent color profile of samples such as blood smear.
“Built-in white-balance of the iPhone 4 is insufficient to overcome variable illumination conditions, resulting in a shift of the apparent color profile of samples such as this blood smear, as quantified in E and F for a region of interest illuminated with a white LED or halogen lamp, respectively.”
Quote in the paperSubstantial claim - Findingown data
For a red plastic film over a pinhole array, the apparent ratio of red-to-green light transmission through the pinholes changes from 10:1 to 1:1 as different color channels saturate, resulting in loss of color contrast.
“The result is loss of color contrast as the apparent ratio of red-to-green light transmission through the pinholes changes from 10∶1 to 1∶1 as different color channels saturate (Figure 6C).”
Quote in the paperSubstantial claim - Findingown data
Phone automation during image capture and storage degrades feature and color information.
“Phone automation during image capture and storage degrades feature and color information.”
Quote in the paperSubstantial claim - Findingnone given
Automated algorithms control focus, exposure, color balance, and image processing in modern mobile phones, significantly altering image appearance from the raw intensity values.
“In contrast, automated algorithms control focus, exposure, color balance, and image processing in modern mobile phones (Figure 5C), significantly altering image appearance from the raw intensity values.”
Quote in the paperSubstantial claim- Rests on an assumption: The default camera application prioritizes a simple image capture process and high perceived sharpness over the reproducibility and detailed control desired by scientific users.
- Findingnone given
Modern mobile phones automatically implement image processing, including image compression and edge sharpening, once images have been collected.
“Once images have been collected, modern mobile phones automatically implement image processing, including image compression and edge sharpening.”
Quote in the paperSubstantial claim - Findingown data
Sharpening algorithms cause patterns of artificial ringing and haloing around high contrast structures.
“Sharpening algorithms cause patterns of artificial ringing and haloing around high contrast structures such as this chrome-on-glass resolution target.”
Quote in the paperSubstantial claim - Findingown data
The procedure resulted in excellent sensor linearity with intensity (R2 value of 0.999).
“This procedure resulted in excellent sensor linearity with intensity (R2 value of 0.999; Figure 3C).”
Quote in the paperSubstantial claim - Interpretationown data
Mobile phone microscopes can provide reliable and repeatable images suitable for diagnostic use.
“The results of our study indicate that mobile phone microscopes can indeed provide reliable and repeatable images suitable for diagnostic use.”
Quote in the paperSubstantial claim - Interpretationreasoning
By accommodating the differences between mobile phone cameras and scientific cameras, mobile phone microscopes can be reliably used to increase access to quantitative imaging for a variety of medical and scientific applications.
“By accommodating the differences between mobile phone cameras and the scientific cameras, mobile phone microscopes can be reliably used to increase access to quantitative imaging for a variety of medical and scientific applications.”
Quote in the paperSubstantial claim- Rests on an assumption: Scientific cameras used for quantitative optical imaging allow independent control of exposure time and color balance and provide access to raw image data free from post-processing.
- Interpretationown data
A 5 megapixel mobile phone with a typical ∼4 mm focal length and f/2.8 lens is capable of capturing the resolution information collected by a representative low-cost microscope system using a 20x eyepiece.
“We find that the iPhone 4 and scientific camera agree well for most objectives, indicating that a 5 megapixel mobile phone with a typical ∼4 mm focal length and f/2.8 lens is capable of capturing the resolution information collected by a representative low-cost microscope system using a 20x eyepiece.”
Quote in the paperSubstantial claim - Premisecitation
The mobile phone represents and displays intensities according to an sRGB color space.
“While surprising from the perspective of scientific imaging, where intensity in an image is proportional to the recorded pixel value, the mobile phone represents and displays intensities according to an sRGB color space [27].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [27] Ebner M (2007) Color Constancy. 1st ed. West Sussex, England: John Wiley and Sons. source
Not checked in the free run.
- Interpretationreasoning
Procedures can avoid the barriers to quantitative imaging caused by automatic focus, exposure, and color gain on mobile phones.
“We find that automatic focus, exposure, and color gain standard on mobile phones can degrade image resolution and reduce accuracy of color capture if uncorrected, and we devise procedures to avoid these barriers to quantitative imaging.”
Quote in the paperSubstantial claim - Premisecitation
This degree of distortion is reasonable for a system employing a microscope eyepiece, which typically introduces pincushion distortion on the order of 3-12%.
“This degree of distortion is reasonable for a system employing a microscope eyepiece, which typically introduces pincushion distortion on the order of 3–12% [26].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [26] Smith WJ (2008) Modern Optical Engineering. 4th ed. McGraw-Hill.
Not checked in the free run.
- Premisecitation
Demosaicing, noise reduction, edge sharpening, and image compression are automatically carried out on most mobile phone images without input from the user.
“Furthermore, demosaicing, noise reduction, edge sharpening, and image compression are automatically carried out on most mobile phone images without input from the user [15].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [15] Omnivision (2012) OV8830 8-megapixel product brief.
Not checked in the free run.
- Interpretationown data
Artifacts from automatic camera parameter adjustments can compromise diagnostic imaging.
“We demonstrate artifacts that result in nonlinear response to input signals, variation in spectral sensitivity, and changes in effective magnification – all of which can compromise diagnostic imaging – and we outline approaches to achieve more reproducible and consistent imaging.”
Quote in the paperSubstantial claim - Premisecitation
Automatic image-processing algorithms may not represent an optimal use of the available pixel information.
“Implementation of these algorithms across phones varies significantly [16], [17], is not typically lossless, and may not represent an optimal use of the available pixel information [18], further complicating attempts to quantify and compare images.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [18] K Hirakawa, 2006, Joint demosaicing and denoising, Image Process IEEE Trans source
Not checked in the free run.
- Interpretationreasoning
The reduction in imaged area by 16x can have a large impact on diagnostic sensitivity.
“This corresponds to a reduction in imaged area by 16x, the square of the magnification, which can have a large impact on diagnostic sensitivity.”
Quote in the paperSubstantial claim - Premisecitation
Automatic image-processing algorithms are not typically lossless.
“Implementation of these algorithms across phones varies significantly [16], [17], is not typically lossless, and may not represent an optimal use of the available pixel information [18], further complicating attempts to quantify and compare images.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [18] K Hirakawa, 2006, Joint demosaicing and denoising, Image Process IEEE Trans source
Not checked in the free run.
- Premisecitation
For an ideal imaging system, resolution is expected to scale with the numerical aperture of the objective and condenser as governed by the Rayleigh criterion for the minimum distinguishable distance between two points.
“For an ideal imaging system, resolution is expected to scale with the numerical aperture (NA) of the objective and condenser as governed by the Rayleigh criterion for the minimum distinguishable distance between two points, δ = 1.22 λ0/(NAobj+NAcond), where λ0 for a broadband white light source is taken as 550 nm, roughly the peak of the human photopic visual response [25], [26].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [25] Inoue S, Spring KR (1997) Video Microscopy: The Fundamentals New York, N.Y.: Plenum Press. 26–44. source
Not checked in the free run.
- cites [26] Smith WJ (2008) Modern Optical Engineering. 4th ed. McGraw-Hill.
Not checked in the free run.
- Premisecitation
Implementation of automatic image-processing algorithms across phones varies significantly.
“Implementation of these algorithms across phones varies significantly [16], [17], is not typically lossless, and may not represent an optimal use of the available pixel information [18], further complicating attempts to quantify and compare images.”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [16] W Kao, 2006, Design considerations of color image processing pipeline for digital cameras, Consum Electron IEEE Trans source
Not checked in the free run.
- cites [17] McKay C, Swaminathan A, Wu M (2008) Image acquisition forensics: Forensic analysis to identify imaging source. 2008 IEEE Int Conf Acoust Speech Signal Process: 1657–1660. source
Not checked in the free run.
- Premisecitation
Diagnosis of malaria species from images of blood smears requires micron-scale resolution of parasite shape.
“For example, diagnosis of malaria species from images of blood smears requires micron-scale resolution of parasite shape [13]”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [13] World Health Organization (2011) Universal Access to Malaria Diagnostic Testing. Geneva: WHO Press.
Not checked in the free run.
- Interpretationreasoning
When mobile phones are used for microscopy, auto-focus can change the total magnification and therefore the size of objects in an image despite unchanged microscope optics.
“This is also true when mobile phones are used for microscopy, where the auto-focus can change the total magnification and therefore size of objects in an image despite unchanged microscope optics.”
Quote in the paperSubstantial claim - Premisecitation
Proper collection of color information is integral to identification of objects in diagnostic medical imaging applications.
“Proper collection of color information is a high priority in the design of imaging pipelines [16], [36] and integral to identification of objects in diagnostic medical imaging applications [37], [38].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [37] K Kayser, 2008, Image standards in tissue-based diagnosis (diagnostic surgical pathology), Diagn Pathol source
Not checked in the free run.
- cites [38] Xiao K, Zardawi F, Yates JM (2012) Colour management system for displaying microscope images. Displays: 1–7. source
Not checked in the free run.
- Premisecitation
Cellphone cameras, like all other commercial cameras meant to be used with standard displays and printing technology, encode image information with a standard sRGB color profile.
“Note that the cellphone cameras, like all other commercial cameras meant to be used with standard displays and printing technology, encode image information with a standard sRGB color profile [27].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [27] Ebner M (2007) Color Constancy. 1st ed. West Sussex, England: John Wiley and Sons. source
Not checked in the free run.
- Interpretationreasoning
Automatic white balancing results in different effective sensitivities for each of the three ranges of wavelengths corresponding to the red, green, and blue components of the color filter array.
“This process results in different effective sensitivities for each of the three ranges of wavelengths corresponding to the red, green, and blue components of the color filter array.”
Quote in the paperSubstantial claim - Premisecitation
DICOM guidelines indicate that a file size reduction of 15-20x for storage can maintain diagnostic quality.
“DICOM guidelines indicate that a file size reduction of 15–20x for storage can maintain diagnostic quality [43], and the high quality storage setting of most phones fulfills this requirement for compliant imaging (Figure 7E).”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [43] NEMA PS3/ISO 12052 (2010) Digital Imaging and Communications in Medicine (DICOM) Supplement 145: Whole Slide Microscopic Image IOD and SOP Classes Rossyln, VA, USA: National Electrical Manufacturers Association.
Not checked in the free run.
- Premisecitation
Automation and additional software that attempts to mimic human perception can significantly and unintentionally alter the reported spectral information in the image.
“This automation, as well as additional software that attempts to mimic human perception, can significantly and unintentionally alter the reported spectral information in the image [36].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [36] G Sharma, 1997, Digital color imaging, Image Process IEEE Trans source
Not checked in the free run.
- Interpretationreasoning
The low compression ratios used during image capture have only negligible contributions to the observed edge distortions as compared to automated sharpening implemented on the mobile phone.
“Though compression could also cause ringing at edges, we do not observe the blocking effect, the most noticeable artifact of the JPEG algorithm [48], and therefore expect that the low compression ratios used during image capture have only negligible contributions to the observed edge distortions as compared to automated sharpening implemented on the mobile phone.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [48] M-Y Shen, 1998, Review of Postprocessing Techniques for Compression Artifact Removal, J Vis Commun Image Represent source
Not checked in the free run.
- Premisecitation
The DICOM standard supports storage of information about illumination conditions and the capture device.
“The Digital Imaging and Communications in Medicine (DICOM) standard, which is a format for exchanging medical image data, supports storage of information about illumination conditions and the capture device [43].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [43] NEMA PS3/ISO 12052 (2010) Digital Imaging and Communications in Medicine (DICOM) Supplement 145: Whole Slide Microscopic Image IOD and SOP Classes Rossyln, VA, USA: National Electrical Manufacturers Association.
Not checked in the free run.
- Interpretationreasoning
As a result, phone-specific artifacts will continue to be visible in mobile phone microscope images, such as those around large, sharp, and high contrast features like edges.
“As a result, phone-specific artifacts will continue to be visible in mobile phone microscope images, such as those around large, sharp, and high contrast features like edges.”
Quote in the paperSubstantial claim - Study design and conductown data
The possibility that ringing at sharp edges was caused by low numerical aperture (coherent) illumination was ruled out by quantifying the edge response of the CellScope mobile phone microscope with two different phones and a configurable CMOS camera.
“While ringing at sharp edges can also be caused by low numerical aperture (coherent) illumination, we ruled out this possibility by quantifying the edge response of the CellScope mobile phone microscope with two different phones, the iPhone 4S and Nexus 4, as well as a configurable CMOS camera with direct access to raw image data.”
Quote in the paperSubstantial claim - Interpretationcitation
If a microscopy application demands discrimination of subtle color differences to measure concentrations or sub-cellular properties, the light source may need to be selected to meet additional metrics for its spectral distribution.
“If a microscopy application demands discrimination of subtle color differences to measure concentrations or sub-cellular properties, the light source may also need to be selected to meet additional metrics for its spectral distribution – an issue of particular concern for solid state lighting solutions [42].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [42] MS Rea, 2008, Color rendering: A tale of two metrics, Color Res Appl source
Not checked in the free run.
- Interpretationreasoning
For applications that rely on the linearity of intensity, such as measuring concentration by absorption or fluorescence, gamma decoding is necessary.
“For applications that rely on the linearity of intensity, such as measuring concentration by absorption or fluorescence, gamma decoding is therefore necessary.”
Quote in the paperSubstantial claim - Interpretationreasoning
A combination of a calibrated phone and light source can be used to collect images as part of a reproducible and DICOM-compliant workflow for color management.
“Thus, a combination of a calibrated phone and light source can be used to collect images as part of a reproducible and DICOM-compliant workflow for color management.”
Quote in the paperSubstantial claim - Interpretationreasoning
When color is used as a contrast mechanism to enable cell counting or morphology evaluation, as in malaria parasites, standardization with a consistent light source may be sufficient.
“When color is used as a contrast mechanism to enable cell counting or morphology evaluation, as in malaria parasites, this type of standardization may be sufficient.”
Quote in the paperSubstantial claim - Interpretationown data
Without applying phone-specific proprietary methods to remove the sharpening, information is lost at sharp, high contrast transitions.
“Without applying phone-specific proprietary methods to remove the sharpening, information is lost at these sharp, high contrast transitions.”
Quote in the paperSubstantial claim - Premisecitation
A scene of punctate samples in dark-field or fluorescence triggers excessive gain and exposure time as the auto-exposure algorithm attempts to achieve a moderate average image intensity.
“This type of scene, as previously noted in the case of video microscopy [41], triggers excessive gain and exposure time as the auto-exposure algorithm attempts to achieve a moderate average image intensity.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [41] Sluder G, Wolf DE, editors (2007) Digital Microscopy, Volume 81: Methods in Cell Biology San Diego, CA: Academic Press. 343–347.
Not checked in the free run.
- Interpretationreasoning
Acquiring color information reproducibly with a mobile phone camera is best accomplished with a consistent light source.
“This means that acquiring color information reproducibly with a mobile phone camera is best accomplished with a consistent light source (Figure 7A), as is the case for a standard clinical microscope with a user-configurable scientific camera.”
Quote in the paperSubstantial claim - Premisecitation
Mobile phones apply automatic processing to an image to increase its apparent sharpness.
“Mobile phones also apply automatic processing to an image to increase its apparent sharpness [34], [47].”
Quote in the paperSubstantial claim0 of 2 sources checked- cites [34] Gonzalez RC, Woods RE (2007) Digital Image Processing Upper Saddle River, NJ: Prentice Hall. 162–165.
Not checked in the free run.
- cites [47] Sharpening and Standardized Sharpening for Comparing Cameras. Available: <ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="http://www.imatest.com/docs/sharpening/" xlink:type="simple">http://www.imatest.com/docs/sharpening/</ext-link>. Accessed 2013 November 27.
Not checked in the free run.
- Premisecitation
An available approach to standardize the sharpening applied at edges assumes a particular symmetrical functional form to the sharpening.
“While an approach to standardize the sharpening applied at edges to compare the underlying imaging devices is available [47], it assumes a particular symmetrical functional form to the sharpening.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [47] Sharpening and Standardized Sharpening for Comparing Cameras. Available: <ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="uri" xlink:href="http://www.imatest.com/docs/sharpening/" xlink:type="simple">http://www.imatest.com/docs/sharpening/</ext-link>. Accessed 2013 November 27.
Not checked in the free run.
- Interpretationreasoning
When an application requires analyzing the intensity of the image instead of displaying it, the gamma encoding needs to be undone to recover the linear response.
“It is only when the application requires analyzing the intensity of the image instead of displaying it that this gamma encoding needs to be undone (thus recovering the linear response shown in the figure).”
Quote in the paperSubstantial claim
- May combine two claims. 24
- The quote only partly states this claim. 1
- Classified as probably not a substantial claim. 18
May combine two claims.
Findingown dataAutomatic focus, exposure, and color gain standard on mobile phones can degrade image resolution and reduce accuracy of color capture if uncorrected.
“We find that automatic focus, exposure, and color gain standard on mobile phones can degrade image resolution and reduce accuracy of color capture if uncorrected, and we devise procedures to avoid these barriers to quantitative imaging.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataAutomatic camera parameter adjustments cause artifacts that result in nonlinear response to input signals, variation in spectral sensitivity, and changes in effective magnification.
“We demonstrate artifacts that result in nonlinear response to input signals, variation in spectral sensitivity, and changes in effective magnification – all of which can compromise diagnostic imaging – and we outline approaches to achieve more reproducible and consistent imaging.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe pixel response is monotonic across the 8-bit range but is not linear.
“We find that the response is monotonic across the 8-bit range but is not linear (Figure 3C).”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataImage linearity, distortion, and color can be corrected as needed in mobile phone microscopy.
“We demonstrate that quantitative microscopy with micron-scale spatial resolution can be carried out with multiple phones and that image linearity, distortion, and color can be corrected as needed.”
Quote in the paperSubstantial claimMay combine two claims.
FindingreasoningExposure time and electronic gain are automatically adjusted on most phones to prevent over- or under-exposure, but this can cause variations in brightness and color response within or between specimens, making comparisons across images difficult.
“Exposure time and electronic gain are automatically adjusted on most phones to prevent over or under-exposure, but this can cause variations in brightness and color response within or between specimens, making comparisons across images by human viewers or by automated analysis software difficult.”
Quote in the paperSubstantial claim- Rests on an assumption: Mobile phone cameras permit limited control of camera parameters and often allow no control over post-processing done on the image before viewing or transmission.
The quote only partly states this claim.
Findingown dataThe mean intensity and distribution of colors of the cells and surrounding area vary noticeably across images taken with different phones.
“Independent of this effect, the mean intensity and distribution of colors of the cells and surrounding area vary noticeably across the images.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe measured spatial resolution of the mobile phone microscope improves with objective NA, as expected theoretically, but is consistently worse than that based on the Rayleigh criterion.
“We find that the measured spatial resolution of our mobile phone microscope improves with objective NA, as expected theoretically, but is consistently worse than that based on the Rayleigh criterion (Figure 2A).”
Quote in the paperSubstantial claim- Rests on an assumption: Features smaller than the measured resolution limit do not have sufficient contrast to be useful for human or computer analysis.
May combine two claims.
Findingown dataPerformance of both Apple and Android phones was similar and dependent primarily on pixel density until the optics of the microscope become limiting.
“Performance of both Apple and Android phones were found to be similar and dependent primarily on pixel density until the optics of the microscope become limiting.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe automated sharpening is neither symmetric nor consistent across mobile phones, as demonstrated by profiles for the Nexus 4 and iPhone 4S.
“As profiles for the Nexus 4 and iPhone 4S demonstrate, however, the automated sharpening is neither symmetric nor consistent across mobile phones.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
Findingnone givenAutomatic white balancing adjusts the gains of the three channels independently to meet pre-specified and often proprietary criteria about the color composition of the scene.
“Automatic white balancing adjusts the gains of the three channels independently to meet pre-specified and often proprietary criteria about the color composition of the scene (Figure 5C).”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
FindingreasoningDevelopers can lock brightness and color gain independently, and have limited control of image manipulation after capture, enabling lossless image saving in place of traditional JPEG compression.
“They also have the ability to lock brightness and color gain independently, as well as limited control of image manipulation after capture, enabling lossless image saving in place of traditional JPEG compression.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Findingnone givenCertain features of the image processing applied by the on-board software cannot be disabled given the degree of control currently provided by phone manufacturers and operating system designers.
“Certain features of the image processing applied by the on-board software cannot be disabled given the degree of control currently provided by phone manufacturers and operating system designers.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim. May combine two claims.
Findingnone givenDevelopers cannot arbitrarily specify gain and exposure values, but they can lock capture settings chosen by the phone's automated processes.
“While developers cannot (yet) arbitrarily specify gain and exposure values, they do have the ability to lock capture settings that are chosen by the phone’s automated processes.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim. May combine two claims.
Findingnone givenMobile phones typically apply a JPEG compression algorithm prior to storage, raising a concern about information that might be lost from and artifacts added to the resulting images.
“Mobile phones typically apply a JPEG compression algorithm prior to storage, raising a concern about information that might be lost from and artifacts added to the resulting images.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe image of the Wright stained blood cells in Figure 2B is 1.7 MB in size when using the highest quality JPEG storage setting, which is a ∼10x reduction from the theoretical 8-bit, 3-color storage size of 15 MB for a 5 megapixel image.
“The image of the Wright stained blood cells in Figure 2B is 1.7 MB in size when using the highest quality JPEG storage setting, which is a ∼10x reduction from the theoretical 8-bit, 3-color storage size of 15 MB for a 5 megapixel image.”
Quote in the paperNot a substantial claimMay combine two claims.
InterpretationreasoningThe lack of measurable resolution increase from 40×/0.65 NA to 60×/0.85 NA is likely due to insufficient aberration correction and to the reduced resolution improvement expected when the objective NA increases beyond the condenser NA.
“Consistent with this, we find that increasing objective numerical aperture and magnification from 40×/0.65 NA to 60×/0.85 NA does not yield a measurable increase in microscope resolution, likely due both to insufficient aberration correction and to the reduced resolution improvement expected when the objective NA increases beyond the condenser NA (Figure 2A).”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
PremisecitationThe speed of diagnosing malignancy on breast biopsy slides improves with consistent color information.
“while the speed of diagnosing malignancy on breast biopsy slides has been shown to improve with consistent color information [14].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [14] E Krupinski, 2012, Observer performance using virtual pathology slides: impact of LCD color reproduction accuracy, J Digit Imaging source
Not checked in the free run.
May combine two claims.
InterpretationreasoningThe resolution of the mobile phone microscope is worse than the Rayleigh criterion due to the low price point of the objectives used ($75-200), with a more significant penalty for high NA objectives.
“We expect that resolution of the mobile phone microscope will be worse than the Rayleigh criterion due to the low price point of the objectives used ($75–200), with a more significant penalty for high NA objectives that are more complex to construct and hence suffer from greater proportional performance degradation at low price points.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Interpretationnone givenThe characterization method described is general to color images when raw pixel image data is unavailable.
“While the characterization method described here is general to color images when raw pixel image data is unavailable, the specific 5 megapixel threshold described corresponds to the particular objective and eyepiece combination presented.”
Quote in the paperNot a substantial claimMay combine two claims.
Study design and conductnone givenThe field of view remained constant for each objective, so the size of the circle could be used to calibrate a sample-referenced pixel pitch for each captured image even as the auto-focus functionality of the phone varied the apparent field size dynamically.
“Since this field of view remained constant for each objective, the size of the circle could be used to calibrate a sample-referenced pixel pitch for each captured image even as the auto-focus functionality of the phone varied the apparent field size dynamically.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationHumans perceive the color of objects to stay approximately constant across illumination conditions with different spectral compositions, but electronic sensors have defined wavelength-dependent sensitivity.
“Humans perceive the color of objects to stay approximately constant across illumination conditions with different spectral compositions [40], but electronic sensors have defined wavelength-dependent sensitivity.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [40] Fairchild MD (2013) Color Appearance Models John Wiley and Sons. 158–159.
Not checked in the free run.
Probably not a substantial claim: may not pass S4 (based).
Premisenone givenCurrent versions of the Android and iOS platforms (4.4 and 7, respectively) enable development of custom camera applications.
“Beyond the basic camera functionality that comes standard with most mobile phones, current versions of the Android and iOS platforms (4.4 and 7, respectively) enable development of custom camera applications.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationGroups conducting cytology-based evaluations have invested significant effort into developing and validating workflows to consistently capture and properly display color information in images.
“While sample preparation itself affects color in the final sample, groups conducting cytology-based evaluations have invested significant effort into developing and validating workflows to consistently capture and properly display color information in images [38].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [38] Xiao K, Zardawi F, Yates JM (2012) Colour management system for displaying microscope images. Displays: 1–7. source
Not checked in the free run.
May combine two claims.
PremisecitationFor images of samples that do not meet pre-set minimum levels of average brightness, the default behavior of mobile phones is to adjust electronic gain and exposure time to meet those minimum brightness levels, which has a significant effect on captured color information.
“For images of samples that do not meet pre-set minimum levels of average brightness [16], the default behavior of mobile phones is to adjust electronic gain and exposure time to meet those minimum brightness levels, which has a significant effect on captured color information.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [16] W Kao, 2006, Design considerations of color image processing pipeline for digital cameras, Consum Electron IEEE Trans source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.
Interpretationnone givenWith the standardization and recording of assay parameters, image data can be structured according to the DICOM standard, enabling the integration of specimen data into existing hospital workflows for picture archival and pathology.
“With the standardization and recording of assay parameters, image data can be structured according to the DICOM standard, enabling the integration of specimen data into existing hospital workflows for picture archival and pathology.”
Quote in the paperNot a substantial claim- Rests on an assumption: To translate the potential of mobile phone microscopy to medical imaging applications, questions about the quality and reproducibility of data captured by a mobile imaging device must first be answered.
May combine two claims.
InterpretationreasoningColor information can be compromised by the phone's automatic white balancing processes and the properties of the light source.
“Color information can also be compromised by the phone’s automatic white balancing processes and the properties of the light source.”
Quote in the paperSubstantial claimClassified as probably not a substantial claim.
PremisecitationStaining dyes transform properties of the sample correlated to pathology into distinct color changes that pathologists and algorithms are trained to recognize.
“Staining dyes transform properties of the sample correlated to pathology into distinct color changes that pathologists, and more recently algorithms [39], are trained to recognize.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [39] J Chang, 2012, Automated Tuberculosis Diagnosis Using Fluorescence Images from a Mobile Microscope, Med Image Comput Comput Interv
Not checked in the free run.
May combine two claims.
InterpretationreasoningWith a fully configurable camera, users can work with a known light source and apply standard settings to avoid wavelength dependence, enabling reproducible image capture.
“With a fully configurable camera, users can work with a known light source and apply standard settings to avoid this wavelength dependence, enabling reproducible image capture.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataAuto-focus motion can accommodate small displacements in the sample plane but is insufficient to fully eliminate the need for a stage focus mechanism on the microscope attachment.
“The auto-focus motion can be useful for accommodating small displacements in the sample plane but is insufficient to fully eliminate the need for a stage focus mechanism on the microscope attachment.”
Quote in the paperSubstantial claimMay combine two claims.
Interpretationown dataThe autofocus-induced change in image scale is not significant for tasks such as counting blood cells, but applications requiring more quantitative information on feature size could be impacted.
“While this effect is not significant for tasks such as counting blood cells, applications that require more quantitative information on feature size could be impacted by changes in the image scale due to autofocus.”
Quote in the paperSubstantial claim- Rests on an assumption: Accurate feature size is often essential for measuring and comparing objects for identification and diagnosis.
Probably not a substantial claim: may not pass S3 (load-bearing).
InterpretationreasoningThe specific 5 megapixel threshold described corresponds to the particular objective and eyepiece combination presented.
“While the characterization method described here is general to color images when raw pixel image data is unavailable, the specific 5 megapixel threshold described corresponds to the particular objective and eyepiece combination presented.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Premisenone givenScientific cameras carry out no compression, sharpening, or any other processing on image data without user instruction.
“In contrast, scientific cameras carry out no compression, sharpening, or any other processing on image data without user instruction.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationMonitoring lung pathology requires identifying changes in cell length and area as a function of disease state.
“For example, monitoring lung pathology requires identifying changes in cell length and area as a function of disease state [35].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [35] M Ochs, 2013, Quantitative microscopy of the lung: a problem-based approach. Part 1: basic principles of lung stereology, Am J Physiol Lung Cell Mol Physiol source
Not checked in the free run.
Classified as probably not a substantial claim. May combine two claims.
InterpretationreasoningThese artifacts do not ultimately affect the basic morphology or color information that is useful for most diagnostic applications but still represent a difference from scientific imaging.
“These artifacts do not ultimately affect the basic morphology or color information that is useful for most diagnostic applications but still represent a difference from scientific imaging.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Study design and conductnone givenPaired with an achromatic 50 mm focal length lens, the reference camera yielded a sample magnification that guaranteed oversampling of a factor of at least ≥4 * Nyquist for the frequency content collected by each objective.
“Paired with an achromatic 50 mm focal length lens, this yielded a sample magnification that guaranteed oversampling of a factor of at least ≥4 * Nyquist for the frequency content collected by each objective.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductnone givenCamera modules were inspected visually for damage before testing.
“Camera modules were inspected visually for damage before testing.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
Study design and conductnone givenThe exposure was locked on the same clear field of view to eliminate any color or exposure based effects.
“The exposure was locked on the same clear field of view to eliminate any color or exposure based effects.”
Quote in the paperSubstantial claimMay combine two claims.
PremisecitationBy the sRGB standard, intensity information is encoded with a gamma factor of 0.45 so that monitors with the inverse gamma factor of 2.2 can linearly display the information.
“By this standard, the intensity information is encoded with a gamma factor of 0.45 as is done by other consumer devices so that monitors with the inverse gamma factor of 2.2 can linearly display the information.”
Quote in the paperSubstantial claim
The paper relies on this without support.
Scientific cameras used for quantitative optical imaging allow independent control of exposure time and color balance and provide access to raw image data free from post-processing.
“Unlike mobile phone cameras, specialized scientific cameras used for quantitative optical imaging applications in medicine allow independent control of parameters such as exposure time and color balance, and they provide access to raw image data free from post-processing.”
The paper relies on this without support.
Mobile phone cameras permit limited control of camera parameters and often allow no control over post-processing done on the image before viewing or transmission.
“Mobile phone cameras, on the other hand, permit limited control of camera parameters and often allow no control over post-processing done on the image before viewing or transmission.”
The paper relies on this without support.
The default camera application prioritizes a simple image capture process and high perceived sharpness over the reproducibility and detailed control desired by scientific users.
“This is further complicated by the default camera application, which prioritizes a simple image capture process and high perceived sharpness over the reproducibility and detailed control desired by scientific users.”
The authors state this as an assumption.
The mobile phone camera sensor uses a common Bayer Filter Array directly on the sensor.
“Since the exact layout of the filter array is proprietary, as is the unique demosaicing algorithm used by each phone [17], we assume a common Bayer Filter Array directly on the sensor.”
The paper relies on this without support.
Features smaller than the measured resolution limit do not have sufficient contrast to be useful for human or computer analysis.
“For most purposes, features smaller than the measured resolution limit do not have sufficient contrast to be useful for human or computer analysis.”
The paper relies on this without support.
Microscope images must have appropriate contrast and consistent colors to be useful for medical diagnostic applications.
“In addition to the need for sufficient spatial resolution, microscope images must have appropriate contrast and consistent colors to be useful for medical diagnostic applications.”
The paper relies on this without support.
Accurate feature size is often essential for measuring and comparing objects for identification and diagnosis.
“Accurate feature size is often essential for measuring and comparing objects for identification and diagnosis.”
The paper relies on this without support.
The potential of mobile phone microscopy has been widely recognized.
“The potential of mobile phone microscopy has been widely recognized.”
The paper relies on this without support.
To translate the potential of mobile phone microscopy to medical imaging applications, questions about the quality and reproducibility of data captured by a mobile imaging device must first be answered.
“To translate this potential to medical imaging applications, however, we must first answer questions about the quality and reproducibility of data captured by a mobile imaging device.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Mobile-based light microscopy offers a cheaper and more transportable tool for diagnosing infectious and haematological conditions.
“Mobile-based light microscopy offers a cheaper and more transportable tool for diagnosing infectious and haematological conditions (70, 73).”
From Assessing the impact of mHealth interventions in low- and middle-income countries – what has been shown to work? (2014) · cites this paper as [73]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Spatial-frequency formula contradicts stated wavelength dependence
MinorThe text says the highest spatial frequency is inversely proportional to wavelength, but its stated formula for that frequency, k = λ/2 NA, makes k proportional to wavelength.
“As with the Rayleigh resolution criterion, this value is proportional to the numerical aperture and inversely proportional to the effective wavelength [25].”
“Imaging without aliasing requires, per the Nyquist criterion, sampling at twice the maximum spatial frequency k captured by the system (k = λ/2 NA for fully incoherent illumination) [28] and thus sets an upper limit on the sample-referenced pixel pitch.”
The citation lists haven’t been fetched yet
References 42 · Cited by 196
Research works printed but not linked: 5
- World Health Organization (2011) Universal Access to Malaria Diagnostic Testing. Geneva: WHO Press.
- Smith WJ (2008) Modern Optical Engineering. 4th ed. McGraw-Hill.
- Gonzalez RC, Woods RE (2007) Digital Image Processing Upper Saddle River, NJ: Prentice Hall. 162–165.
- Fairchild MD (2013) Color Appearance Models John Wiley and Sons. 158–159.
- Sluder G, Wolf DE, editors (2007) Digital Microscopy, Volume 81: Methods in Cell Biology San Diego, CA: Academic Press. 343–347.
- Authors
- Skandarajah, A., Reber, C. D., Switz, N. A., & Fletcher, D. A.
- Affiliations
- University of California, Berkeley (US)
- Lawrence Berkeley National Laboratory (US)
- Corresponding author
- Fletcher, Daniel A.
- Venue
- PLoS ONE
- Publisher
- Public Library of Science (PLoS)
- Published
- 2014-05-13
- Volume
- 9
- Issue
- 5
- Pages
- e96906
- ISSN
- 1932-6203 (ISSN-L)
- PMC
- PMC4019540
- OpenAlex
- W2024363851
- Cited by
- 196
- References
- 42
- Citations by year
- 2026: 8 · 2025: 7 · 2024: 11 · 2023: 14 · 2022: 5 · 2021: 17 · 2020: 17 · 2019: 26 · 2018: 25 · 2017: 23 · 2016: 18 · 2015: 21 · 2014: 4
- FWCI
- 33.54
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- PLoS ONE (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, other-oa)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) (submittedVersion)
- Funders
- National Science Foundation
- Bill and Melinda Gates Foundation
- Intel Corporation
- Microsoft Research
- University of California Berkeley
- Blum Center for Developing Economies, University of California Berkeley
- Topics
- Physical Sciences › Engineering › Media Technology › Image Processing Techniques and Applications
- Physical Sciences › Engineering › Biomedical Engineering › Advanced optical system design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › CCD and CMOS Imaging Sensors
- Keywords
- quantitative microscopy, quantitative image analysis, diffraction-limited resolution, single-cell imaging, iPhone, Android smartphones, image distortion correction
- MeSH terms
- Diagnostic Imaging, Humans, Image Processing, Computer-Assisted, Microscopy, Photography, Cell Phone
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
