GPU Accelerated Generation of Digitally Reconstructed Radiographs for 2-D/3-D Image Registration
The DOI matches its Crossref record; 39 of the 42 references it cites were found in a registry or an index.
Abstract
Recent advances in programming languages for graphics processing units (GPUs) provide developers with a convenient way of implementing applications which can be executed on the CPU and GPU interchangeably. GPUs are becoming relatively cheap, powerful, and widely available hardware components, which can be used to perform intensive calculations. The last decade of hardware performance developments shows that GPU-based computation is progressing significantly faster than CPU-based computation, particularly if one considers the execution of highly parallelisable algorithms. Future predictions illustrate that this trend is likely to continue. In this paper, we introduce a way of accelerating 2-D/3-D image registration by developing a hybrid system which executes on the CPU and utilizes the GPU for parallelizing the generation of digitally reconstructed radiographs (DRRs). Based on the advancements of the GPU over the CPU, it is timely to exploit the benefits of many-core GPU technology by developing algorithms for DRR generation. Although some previous work has investigated the rendering of DRRs using the GPU, this paper investigates approximations which reduce the computational overhead while still maintaining a quality consistent with that needed for 2-D/3-D registration with sufficient accuracy to be clinically acceptable in certain applications of radiation oncology. Furthermore, by comparing implementations of 2-D/3-D registration on the CPU and GPU, we investigate current performance and propose an optimal framework for PC implementations addressing the rigid registration problem. Using this framework, we are able to render DRR images from a 256×256×133 CT volume in ~24 ms using an NVidia GeForce 8800 GTX and in ~2 ms using NVidia GeForce GTX 580. In addition to applications requiring fast automatic patient setup, these levels of performance suggest image-guided radiation therapy at video frame rates is technically feasible using relatively low cost PC architecture.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“3B görüntüler üzerinde GPU ile 50 kat hızlanma sağlanmıştır.”
From GPU Programlamada CUDA Platformu Kullanılan Paralel Görüntü İşleme Çalışmalarının İncelenmesi (2020) · cites this paper as [73]
Citations
Cites
Principles of Computerized Tomographic Imaging
Kak, A. C., & Slaney, M. (2001) · doi:10.1137/1.9780898719277 · Cited by 5,809
Programming Massively Parallel Processors: A Hands-on Approach
Kirk, D. B., & Hwu, W. (2012) · Cited by 1,801
Fast calculation of the exact radiological path for a three‐dimensional CT array
Siddon, R. L. (1985) · Medical Physics · doi:10.1118/1.595715 · Cited by 1,535
Survey: interpolation methods in medical image processing
Lehmann, T. M., Gonner, C., & Spitzer, K. (1999) · IEEE Transactions on Medical Imaging · doi:10.1109/42.816070 · Cited by 1,340
Fast volume rendering using a shear-warp factorization of the viewing transformation
Lacroute, P., & Levoy, M. (1994) · Proceedings of the 21st annual conference on Computer graphics and interactive techniques - SIGGRAPH '94 · doi:10.1145/192161.192283 · Cited by 1,013
Footprint evaluation for volume rendering
Westover, L. (1990) · Proceedings of the 17th annual conference on Computer graphics and interactive techniques · doi:10.1145/97879.97919 · Cited by 787
A Fast Voxel Traversal Algorithm for Ray Tracing
Amanatides, J., & Woo, A. (1987) · doi:10.2312/egtp.19871000 · Cited by 660
Programming Massively Parallel Processors. A Hands-on Approach
Cheng, J. (2010) · Scalable Computing Practice and Experience · doi:10.12694/scpe.v11i3.654 · Cited by 654
Nguyen, H. (2007) · Cited by 597
Acceleration techniques for GPU-based volume rendering
Kruger, J., & Westermann, R. (2004) · IEEE Transactions on Ultrasonics, Ferroelectrics and Frequency Control · doi:10.1109/visual.2003.1250384 · Cited by 312
Research works printed but not linked: 1
- nguyen, 2007, GPU Gems 3
Cited by
Medical image processing on the GPU – Past, present and future
Eklund, A., Dufort, P., Forsberg, D., & LaConte, S. M. (2013) · Medical Image Analysis · doi:10.1016/j.media.2013.05.008 · Cited by 397
“Dorgham et al. (2012) proposed faster rendering by sparse sampling.”
2D/3D Multimode Medical Image Registration Based on Normalized Cross-Correlation
Liu, S., Yang, B., Wang, Y., Tian, J., Yin, L., & Zheng, W. (2022) · Applied Sciences · doi:10.3390/app12062828 · Cited by 137
Montúfar, J., Romero, M., & Scougall‐Vilchis, R. J. (2018) · American Journal of Orthodontics and Dentofacial Orthopedics · doi:10.1016/j.ajodo.2017.06.028 · Cited by 86
Otake, Y., Wang, A. S., Webster Stayman, J., Uneri, A., Kleinszig, G., Vogt, S., Khanna, A. J., Gokaslan, Z. L., & Siewerdsen, J. H. (2013) · Physics in Medicine and Biology · doi:10.1088/0031-9155/58/23/8535 · Cited by 89
“For example: Otake et al (2012c) reported ~500 eval s −1 , Fisher et al (2013) reported ~400 eval s −1 (DRR rendering speed), Dorgham et al (2012) reported ~167 eval s −1 , Varnavas et al (2013) reported ~21.5 eval s −1 , Steininger et al (2012) reported ~13.3 eval s −1 , and Spoerk et al (2012)…”
BackgroundInfluentialMontúfar, J., Romero, M., & Scougall‐Vilchis, R. J. (2018) · American Journal of Orthodontics and Dentofacial Orthopedics · doi:10.1016/j.ajodo.2017.08.028 · Cited by 73
Reyneke, C., Lüthi, M., Burdin, V., Douglas, T. S., Vetter, T. R., & Mutsvangwa, T. (2018) · IEEE Reviews in Biomedical Engineering · doi:10.1109/rbme.2018.2876450 · Cited by 72
“tration for purposes such as preinterventional patient alignment (image-guided registration therapy) [100].”
BackgroundKim, M., Liu, Y., Oh, S. H., Ahn, H., Kim, S., & Nelson, G. (2021) · Sensors · doi:10.3390/s21020505 · Cited by 56
“A 2D radiograph synthesized from the CBCT has the advantages of maintaining the object size without magnification, and the ability to adjust the head position to reduce distortion (reorientation) [8–11].”
BackgroundEnhancing the Security of Exchanging and Storing DICOM Medical Images on the Cloud
Dorgham, O. M., Al-Rahamneh, B., Almomani, A., Al-Hadidi, M. R., & Khatatneh, K. (2017) · International Journal of Cloud Applications and Computing · doi:10.4018/ijcac.2018010108 · Cited by 44
Self-citationIncreasing the Automation of a 2D-3D Registration System
Varnavas, A., Carrell, T., & Penney, G. (2013) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2012.2227337 · Cited by 34
Dhont, J., Verellen, D., Mollaert, I., Vanreusel, V., & Vandemeulebroucke, J. (2020) · Radiotherapy and Oncology · doi:10.1016/j.radonc.2020.10.004 · Cited by 37
Whitmarsh, T., Humbert, L., Barquero, L. M. D. R., Gregorio, S. D., & Frangi, A. F. (2013) · Medical Image Analysis · doi:10.1016/j.media.2013.02.002 · Cited by 28
X-ray system simulation software tools for radiology and radiography education
Kengyelics, S. M., Treadgold, L. A., & Davies, A. G. (2017) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2017.12.005 · Cited by 26
Bidirectional elastic image registration using B-spline affine transformation
Gu, S., Meng, X., Sciurba, F. C., Ma, H., Leader, J. K., Kaminski, N., Gur, D., & Pu, J. (2014) · Computerized Medical Imaging and Graphics · doi:10.1016/j.compmedimag.2014.01.002 · Cited by 27
“There have been numerous algorithms [1–3] developed for this purpose in the past, particularly in the area of medical image analysis [4–7].”
MethodRegistration for Orthopaedic Interventions
Yaniv, Z. (2015) · Lecture notes in computational vision and biomechanics · doi:10.1007/978-3-319-23482-3_3 · Cited by 24
“Currently this is no longer an issue, as the use of high performance Graphical Processing Units (GPU) has become commonplace and efficient creation of DRRs on the GPU is both fast and cost effective [29, 126].”
BackgroundDuménil, A., Kaladji, A., Castro, M., Göksu, C., Lucas, A., & Haigron, P. (2016) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-016-1416-1 · Cited by 27
“Some works have focused on accelerating the generation of DRRs by opting for fast CPU computing strategies, with [23] or without pre-calculation [24], then for GPU computing strategies [25,26].”
BackgroundBrain big data processing with massively parallel computing technology: challenges and opportunities
Chen, D., Hu, Y., Cai, C., Zeng, K., & Li, X. (2016) · Software Practice and Experience · doi:10.1002/spe.2418 · Cited by 27
“It utilized the GPU to parallelize the generation of digitally reconstructed radiographs [43].”
MethodIterative Closest Graph Matching for Non-rigid 3d/2d Coronary Arteries Registration
Zhu, J., Li, H., Ai, D., Yang, Q., Fan, J., Huang, Y., Song, H., Han, Y., & Yang, J. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2020.105901 · Cited by 22
Intensity-based 2D 3D registration for lead localization in robot guided deep brain stimulation
Hunsche, S., Sauner, D., Majdoub, F. E., Neudorfer, C., Poggenborg, J., Goßmann, A., & Maarouf, M. (2017) · Physics in Medicine and Biology · doi:10.1088/1361-6560/aa5ecd · Cited by 20
Ghafurian, S., Hacihaliloglu, I., Metaxas, D., Tan, V., & Li, K. (2016) · Neurocomputing · doi:10.1016/j.neucom.2016.07.070 · Cited by 20
“The intensity value of each pixel in the DRR is the integral of the volume intensities that the ray from the ray-source to that pixel has passed through [18].”
BackgroundFast reconstructed radiographs from octree-compressed volumetric data
Fisher, M. H., Dorgham, O. M., & Laycock, S. D. (2012) · International Journal of Computer Assisted Radiology and Surgery · doi:10.1007/s11548-012-0783-5 · Cited by 19
“Our RCT implementation [8] runs on the GPU and takes advantage of NVIDIATM CUDATM architecture [33].”
MethodSelf-citationNaik, R. R., Hoblidar, A., Bhat, S. N., Ampar, N., & Kundangar, R. (2022) · Journal of Imaging · doi:10.3390/jimaging8070185 · Cited by 16
“where I is the projected pixel intensity through the tissue volume, I0 is the X-ray intensity of X-ray photon directly received by the detector without undergoing any degree of attenuation, μ(ECT) is the LAC of tissue voxel—computed relative to water attenuation (μwater(ECT)) at CT energy (ECT) from Hounsfield Unit(HU…”
MethodDorgham, O. M., Al-Mherat, I., Alshaer, J. J., Bani‐Ahmad, S., & Laycock, S. D. (2019) · Future Internet · doi:10.3390/fi11010025 · Cited by 14
“Different applications could benefit from the results of this research by applying artificial neural network as in [14,15].”
Self-citationHeuristic tree searching for pose-independent 3D/2D rigid registration of vessel structures
Zhu, J., Fan, J., Guo, S., Ai, D., Song, H., Wang, C., Zhou, S., & Yang, J. (2020) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ab6b43 · Cited by 14
“Hipwell et al (2003) analyzed six similarity metrics between DRR/MIP and x-ray images, in which pattern intensity and gradient difference exhibit the best performance.”
MethodLevine, L., & Levine, M. S. (2020) · Journal of Clinical Imaging Science · doi:10.25259/jcis_105_2020 · Cited by 14
Naik, R. R., Anitha, H., Bhat, S. N., Ampar, N., & Kundangar, R. S. (2022) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-022-02600-5 · Cited by 17
“1 3 Where “I” is the projected pixel intensity through tissue volume, “ Io ” is the X-ray intensity subjected to no attenuation, and “l” is the voxel intersection length and “i” denotes voxel index [36].”
BackgroundMethodAutomatic body segmentation for accelerated rendering of digitally reconstructed radiograph images
Dorgham, O. M., Ryalat, M. H., & Naser, M. A. (2020) · Informatics in Medicine Unlocked · doi:10.1016/j.imu.2020.100375 · Cited by 16
Self-citation3D/2D image registration using weighted histogram of gradient directions
Ghafurian, S., Hacihaliloglu, I., Metaxas, D., Tan, V., & Li, K. (2015) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2081316 · Cited by 13
“This problem has usually been address by using alternate ways to either estimate or precalculate the DRRs, which come at the cost of quality or memory and pre-calculation time.(2,3) Therefore, the development of a registration method with minimal number of required DRRs is in demand.”
Method3D/2D Vessel Registration Based on Monte Carlo Tree Search and Manifold Regularization
Zhu, J., Wang, C., Zhang, Y., Zhan, M., Zhao, W. G., Teng, S., Lu, L., & Teng, G. (2023) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2023.3347896 · Cited by 14
“Digitally reconstructed radiographs (DRR) [23] and maximum intensity projection (MIP) [24] are standard methods that generate simulated X-ray projections of computed tomography (CT) images.”
MethodGPU-accelerated Kernel Regression Reconstruction for Freehand 3D Ultrasound Imaging
Wen, T., Li, L., Zhu, Q., Qin, W., Gu, J., Yang, F., & Xie, Y. (2017) · Ultrasonic Imaging · doi:10.1177/0161734616689464 · Cited by 11
Ghafurian, S., Metaxas, D., Tan, V., & Li, K. (2016) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2217756 · Cited by 10
“The traditional method for the generation of DRRs is ray-casting.(1) In this method, the value of each pixel in the DRR is determined by sampling and averaging intensity values along the ray connecting the ray source to that pixel (Figure 1).”
BackgroundMethodInfluentialDigitally Reconstructed Radiographs
(2015) · Series in medical physics and biomedical engineering · doi:10.1201/b18968-3 · Cited by 8
Segmentation of Medical X-ray Bone Image Using Different Image Processing Techniques
Isinkaye, F. O., Aluko, A. G., & Jongbo, O. A. (2021) · International Journal of Image Graphics and Signal Processing · doi:10.5815/ijigsp.2021.05.03 · Cited by 8
“…physician in proffering solution to patient’s ailment. medical image processing is an important area of interest in health care industries based on the fact that majority of medical images contains vital information about patients cases which tends to improve visualization of human anatomy [5,6].”
BackgroundA Fast DRR Generation Scheme for 3D-2D Image Registration Based on the Block Projection Method
Mu, Z. (2016) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw.2016.82 · Cited by 11
“Considering the typical speed-up factors by GPU over equivalent CPU-based implementation were reported to be greater than 50 except for small CT datasets [15], the computing time may be further reduced to sub-millisecond level, making real time or near real time registration possible.”
BackgroundGPU accelerated dynamic respiratory motion model correction for MRI-guided cardiac interventions
Xu, R., & Wright, G. A. (2016) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2016.08.003 · Cited by 6
Automatic body segmentation from computed tomography image
Dorgham, O. M. (2017) · International Conference on Advanced Technologies for Signal and Image Processing · doi:10.1109/atsip.2017.8075612 · Cited by 6
“Entropy is used in different applications of image processing, such as segmentation, classification, image retrieval or registration [20][21].”
MethodSelf-citationCAHECA: Computer Aided Hepatocellular Carcinoma therapy planning
Adeshina, A. M., Hashim, R., & Khalid, N. E. A. (2014) · Interdisciplinary Sciences Computational Life Sciences · doi:10.1007/s12539-013-0204-7 · Cited by 6
“D image registration proposed by Dorgham et al. (2012) would have been a breakthrough in volume visualization because it focuses on approximate methods which reduce the computational overhead while not neglecting the acceptable clinical image quality of datasets, but the authors did not make any…”
MethodMiao, S., Huynh, T., Adnet, C., Pfister, M., & Liao, R. J. (2013) · Lecture notes in computer science · doi:10.1007/978-3-642-40843-4_10 · Cited by 5
Yang, K., Luo, Y., Zhao, Y., Su, S., Qu, D., Zhao, X., & Song, G. (2021) · Physics in Medicine and Biology · doi:10.1088/1361-6560/abe9f5 · Cited by 5
GPU Programlamada CUDA Platformu Kullanılan Paralel Görüntü İşleme Çalışmalarının İncelenmesi
AYDIN, S., SAMET, R., & BAY, Ö. F. (2020) · Politeknik Dergisi · doi:10.2339/politeknik.563767 · Cited by 3
“registration using GPU”, Machine Vision and Applications, 22: 837-850, (2011) [75] Dorgham O.”
MethodYan, H., & Dai, J. (2022) · Mathematics · doi:10.3390/math11010069 · Cited by 3
Bhat, V., Bhat, S. N., & Anitha, H. (2017) · International Conference on Advanced Computing · doi:10.1109/icoac.2017.8441330 · Cited by 3
“Only a subset of the entire rayswas used; missing pixel values were calculated using nearest neighbor interpolation [13].”
MethodProposed Method for Automatic Segmentation of Medical Images
Dorgham, O. M., Nasser, M. A., Ryalat, M. H., & Almomani, A. (2018) · International Arab Conference on Information Technology (ACIT) · doi:10.1109/acit.2018.8672688 · Cited by 3
“So, these images play an important role in the image-guided therapy and radiation oncology [16][17].”
BackgroundSelf-citationA novel fast DRR generation method by reassembling previously generated projection blocks
Mu, Z. (2016) · IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI ...) · doi:10.1109/bhi.2016.7455887 · Cited by 2
Research on Programming Model and Compilation Optimization Technology of Multi-Core GPU
Chen, L. (2022) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2173/1/012080 · Cited by 2
“In 2007, NVIDIA introduced its new GPU architecture CUDA (Compute Unified Device Architecture), the latest support CUDA architecture GPU contains 1024 computing cores, which has been put into practical use, but how efficient use of hundreds of computing core is still facing challenge [3].”
BackgroundReal-Time Simulation of Radiological Images Using CUDA Technology
Gianaria, E., & Gallio, E. (2015) · Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP) · doi:10.1109/pdp.2015.55 · Cited by 1
“[5] have considerably improved computation time of 2D to 3D registration algorithms, parallelizing on GPU the generation of digitally reconstructed radiographs (DRRs).”
BackgroundShu, L., Li, M., Guo, X. J., Chen, Y., Pu, X., & Lin, C. (2024) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ad450a · Cited by 2
Oh, J., & Koo, S. (2024) · Journal of medical imaging · doi:10.1117/1.jmi.11.3.033503 · Cited by 2
An Improved Security Approach for Medical Images and Patients’ Information Transmission
Hafsa, A., Malek, J., & Machhout, M. (2022) · International Conference on Sciences of Electronics, Technologies of Information and Telecommunications · doi:10.1109/setit54465.2022.9875481 · Cited by 1
“The Telemedicine application is a one kind of application that eases the transmission of D-ICOM files in networks [1,2].”
BackgroundCUDA-Accelerated Volume Rendering of Two-phase Bone
Liu, T., Ming-jun, L., Xiuyang, Z., Li, Z., & Jing, W. (2013) · International Conference on Intelligent System Design and Engineering Application (ISDEA) · doi:10.1109/isdea.2012.125
“Dorgham [19]propose an optimal framework for PC implementations addressing the rigid registration problem.”
BackgroundMulti-GPU Accelerated Parallel Algorithm of Wallis Transformation for Image Enhancement
Xiao, H., Song, Y.-P., & Zhou, Q. (2014) · International Journal of Grid and Distributed Computing · doi:10.14257/ijgdc.2014.7.2.10 · Cited by 2
“GPU has possessed powerful parallel computational capability [6].”
BackgroundFast parallel volume visualization on CUDA technology
Adeshina, A. M. (2013) · UTHM Institutional Repository (Universiti Tun Hussein Onn Malaysia)
Ghafurian, S., Hacihaliloglu, I., Metaxas, D., Tan, V., & Li, K. (2017) · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · doi:10.1117/12.2254678 · Cited by 2
Automated and computationally efficient joint motion analysis using low quality fluoroscopy images
Ghafurian, S. (2017) · doi:10.7282/t3s46vdm
“This problem has usually been addressed by using alternate ways to either estimate or precalculate the DRRs, which come at the cost of quality, or memory and precalculation time [19, 56, 58, 10].”
BackgroundMethodInfluentialA GPGPU-Based Algorithm Acceleration System for ECG Signal Processing
Huang, T., Wang, Z., Hu, G. H., Pun, S. H., Li, H.-C., & Vai, M. I. (2024) · International Conference on Algorithms, Computing and Artificial Intelligence (ACAI) · doi:10.1109/acai63924.2024.10899701
“Dorgham et al. proposed a CUDA-based approach that utilizes GPU acceleration to generate digitally reconstructed radiographs (DRRs) for 2D/3D registration [11].”
Fusion d'images et de modèles pour le guidage d'interventions endovasculaires
Duménil, A. (2015) · theses.fr (ABES) · doi:10.70675/c51c7964zaf2ez48b8za186za930ede89a7b · Cited by 3
A GPU-Based Virtual CT Scan System for Data Augmentation
Akase, T., Higaki, T., Manabe, T., & Matsutomo, S. (2026) · IIEEJ International Conference on Image Electronics and Visual Computing (IEVC) · doi:10.1109/ievc69170.2026.11508514
Metadata
- Authors
- Dorgham, O. M., Laycock, S. D., & Fisher, M. H.
- Affiliations
- Al-Balqa Applied University (JO)
- University of East Anglia (GB)
- Venue
- IEEE Transactions on Biomedical Engineering
- Published
- 2012-09
- Volume
- 59
- Issue
- 9
- Pages
- 2594-2603
- ISSN
- 0018-9294 (ISSN-L)
- 1558-2531
- OpenAlex
- W2027752865
- Cited by
- 56
- References
- 42
- Citations by year
- 2026: 1 · 2024: 3 · 2023: 1 · 2022: 6 · 2021: 3 · 2020: 5 · 2019: 2 · 2018: 4 · 2017: 8 · 2016: 7 · 2015: 5 · 2014: 3 · 2013: 7 · 2012: 1
- FWCI
- 4.94
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Biomedical Engineering (publishedVersion)
- PubMed (publishedVersion)
- UEA Digital Repository (University of East Anglia) (acceptedVersion)
- Funders
- Nvidia
- Topics
- Physical Sciences › Physics and Astronomy › Radiation › Advanced Radiotherapy Techniques
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Computer Graphics and Visualization Techniques
- Keywords
- digitally reconstructed radiographs, GPU acceleration, CPU-GPU hybrid computing, rigid registration, image-guided radiotherapy, radiation oncology, video frame rate, parallel computing
- MeSH terms
- Algorithms, Computer Graphics, Humans, Lung, Pelvis, Radiographic Image Enhancement, Imaging, Three-Dimensional
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.
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.