Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis
Open accessAbstract
Computer vision, biomedical image processing and deep learning are related fields with a tremendous impact on the interpretation of medical images today. Among biomedical image sensing modalities, ultrasound (US) is one of the most widely used in practice, since it is noninvasive, accessible, and cheap. Its main drawback, compared to other imaging modalities, like computed tomography (CT) or magnetic resonance imaging (MRI), consists of the increased dependence on the human operator. One important step toward reducing this dependence is the implementation of a computer-aided diagnosis (CAD) system for US imaging. The aim of the paper is to examine the application of contrast enhanced ultrasound imaging (CEUS) to the problem of automated focal liver lesion (FLL) diagnosis using deep neural networks (DNN). Custom DNN designs are compared with state-of-the-art architectures, either pre-trained or trained from scratch. Our work improves on and broadens previous work in the field in several aspects, e.g., a novel leave-one-patient-out evaluation procedure, which further enabled us to formulate a hard-voting classification scheme. We show the effectiveness of our models, i.e., 88% accuracy reported against a higher number of liver lesion types: hepatocellular carcinomas (HCC), hypervascular metastases (HYPERM), hypovascular metastases (HYPOM), hemangiomas (HEM), and focal nodular hyperplasia (FNH).
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
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?
Authors in [102] reported a power efficiency of 1.64 GOps/s/W and a reasonable accuracy of 78.4% including the implementation of CMSIS-NN.
“The authors reported a power efficiency of 1.64 GOps/s/W achieving a reasonable accuracy of 78.4% (including the implementation of CMSIS-NN).”
From A Comprehensive Survey on TinyML (2023) · cites this paper as [102]
The Results section describes CEUS liver-lesion classification experiments on a CPU and GPU, and the Discussion reports classification accuracies, but neither reports the claimed power efficiency or 78.4% accuracy. Searches for “1.64 GOps/s/W,” “78.4%,” “CMSIS-NN,” and power-efficiency terms found no passage addressing those figures or a CMSIS-NN implementation. (LLM, conf=0.77)
Experimental — accuracy not yet measuredDetection of cardiac arrhythmia was boosted by the usage of low-powered microcontrollers as indicated by [102].
“The detection process was boosted by the usage of low-powered microcontrollers as indicated by [102].”
From A Comprehensive Survey on TinyML (2023) · cites this paper as [102]
The Introduction discusses small-memory-footprint models as a step toward a medical embedded system, but does not say low-powered microcontrollers were used or improved detection. The Results and Discussion concern diagnosis of focal liver lesions, not cardiac arrhythmia; searches for arrhythmia detection and microcontrollers found no supporting passage in our copy. (LLM, conf=0.78)
Experimental — accuracy not yet measuredIn [102], the team recorded an accuracy drop from 80% to 78.4% as soon as they deployed it on the edge.
“In [102] for instance, the team recorded an accuracy drop from 80% to 78.4% as soon as they deployed it in the edge.”
From A Comprehensive Survey on TinyML (2023) · cites this paper as [102]
The Results (§4) describe experiments on a CPU/GPU workstation, and the Discussion (§5) attributes an accuracy drop from 95.71% to 56% to a change in evaluation procedure. Searches for 78.4% and edge deployment found no passage reporting the claimed 80%-to-78.4% drop. (LLM, conf=0.76)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
The citation lists haven’t been fetched yet
References 37 · Cited by 63
Research works printed but not linked: 2
- [9]Bishop C.M. Pattern Recognition and Machine Learning Springer Berlin/Heidelberg, Germany 2006
- [39]Tan M. Le Q.V. EfficientNet: Rethinking model scaling for conutional neural networks Proceedings of the International Conference on Machine Learning Long Beach, CA, USA 10–15 June 2019
- Authors
- Căleanu, C. D., Sîrbu, C. L., & Simion, G.
- Affiliations
- Politehnica University of Timisoara (RO)
- Corresponding author
- Simion, Georgiana
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2021-06-16
- Volume
- 21
- Issue
- 12
- Pages
- 4126
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC8235629
- OpenAlex
- W3172157563
- Cited by
- 63
- References
- 37
- Citations by year
- 2026: 5 · 2025: 16 · 2024: 14 · 2023: 17 · 2022: 11
- FWCI
- 3.70
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI in cancer detection
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Medical Image Segmentation Techniques
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- contrast-enhanced ultrasound, focal liver lesions, computer-aided diagnosis, deep neural network, hepatocellular carcinoma, liver metastasis, hemangioma, focal nodular hyperplasia, ultrasound image classification
- MeSH terms
- Contrast Media, Carcinoma, Hepatocellular, Humans, Liver, Liver Neoplasms, Magnetic Resonance Imaging, Ultrasonography, Focal Nodular Hyperplasia
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
