Artificial Intelligence in the Diagnosis of Hepatocellular Carcinoma: A Systematic Review
Abstract
:1. Introduction
2. Materials and Methods
2.1. Searches
2.2. Inclusion and Exclusion Criteria
2.3. Quality Assessment
- The number of images, estimating the risk of bias and overfitting: fewer than 50 (score 0), 50 to 100 (score 1), and more than 100 (score 2) [5]. This factor was considered the most frequently reported in articles. Where only the number of patients was reported, we considered at least one image per patient.
- The use of a completely independent cohort for validation: no cohort (score 0), the partition of the cohort between completely separated training and test set (score 1), external validation cohort (score 2).
- By 2011, the speed of graphics processing units had increased significantly, making it possible to train convolutional neural networks “without” the layer-by-layer pre-training. With the increased computing speed, deep learning had significant advantages in terms of efficiency and speed: no data (score 0), before 2011 (score 1), 2011 or after (score 2).
2.4. Study Selection & Data Extraction
- PMID; First author; Year of publication; Country; Journal.
- The number of patients; Diagnostic method; AI method.
- Research question; Key findings.
- Quality score.
3. Results
3.1. Searching Results
3.2. Quality Assessments
3.3. Results
Study | Country/Region | Journal | Total Study Population | Diagnostic Method | Research Question/Purpose | AI Method | Key Findings |
---|---|---|---|---|---|---|---|
Ziegelmayer et al., 2022 [8] | Germany | Investigative Radiology | 60 patients | CT | To compare the robustness of CNN features versus radiomics features to technical variations in image acquisition parameters. | CNN | CNN features were more stable. |
Xu et al., 2022 [9] | China | Computational and Mathematical Methods in Medicine | 211 patients (122 training set, 89 testing set) | CT | To establish an SVM based on radiomic features at non-contrast CT to train a discriminative model for HCC and ICCA at early stage. | SVM | The model may facilitate the differential diagnosis of HCC and ICCA in the future. |
Turco et al., 2022 [10] | USA | IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency control | 72 patients | US | Proposes an interpretable radiomics approach to differentiate between malignant and benign FLLs on CEUS. | Logistic regression, SVM, RF, and KNN | Aspects related to perfusion (peak time and wash-in time), the microvascular architecture (spatiotemporal coherence), and the spatial characteristics of contrast enhancement at wash-in (global kurtosis) and peak (GLCM Energy) are particularly relevant to aid FLLs diagnosis. |
Sato et al., 2022 [11] | Japan | Journal of Gastroenterology and Hepatology | 972 patients (864 training set, 108 testing set) | US | To analyse the diagnostic performance of deep multimodal representation model-based integration of tumour image, patient background, and blood biomarkers for the differentiation of liver tumours observed using B-mode US. | CNN | The integration of patient background information and blood biomarkers in addition to US images, multi-modal representation learning outperformed the CNN model that used US images alone. |
Rela et al., 2022 [12] | India | International Journal of Advanced Technology and Engineering Exploration | 68 patients (51 training set, 17 testing set) | CT | Different machine learning algorithms are used to classify the tumour as liver abscess and HCC. | SVM, KNN, Decision tree, Ensemble, and Naive Bayes | SVM classifier gives better performance compared to all other AI methods in the study. |
Zheng et al., 2021 [13] | China | Physics in Medicine and Biology | 120 patients (56 training set with 5376 images, 64 testing set with 6144 images) | MRI | To investigate the feasibility of automatic detection of small HCC (≤2 cm) based on PM-DL model. | CNN | The superior performance both in the validation cohort and external test cohort indicated the proposed PM-DL model may be feasible for automatic detection of small HCCs with high accuracy. |
Yang et al., 2021 [14] | Taiwan | PLoS One | 731 patients (394 training set with 10,130 images, 337 testing set with 22,116 images) | CT | To use a previously proposed mask region–based CNN for automatic abnormal liver density detection and segmentation based on HCC CT datasets from a radiological perspective. | CNN | The study revealed that this single deep learning model cannot replace the complex and subtle medical evaluations of radiologists, but it can reduce tedious labour. |
Stollmayer et al., 2021 [15] | Hungary | World Journal of Gastroenterology | 69 patients (training set with 186 images, testing set with 30 images) | MRI | To compare diagnostic efficiency of 2D and 3D-densely connected CNN (DenseNet) for FLLs on multi-sequence MRI. | CNN | Both 2D and 3D-DenseNets can differentiate FNH, HCC and MET with good accuracy when trained on hepatocyte-specific contrast-enhanced multi-sequence MRI volumes. |
Kim et al., 2021 [16] | Korea | European Radiology | 1320 patients (training set with 568 images, testing set with 589 images, tuning set with 193 images) | CT | To develop and evaluate a deep learning-based model capable of detecting primary hepatic malignancies in multiphase CT images of patients at high risk for HCC. | CNN | The proposed model exhibited an 84.8% of sensitivity with 4.80 false positives per CT scan in the test set. |
Căleanu et al., 2021 [17] | Romania | Sensors | 91 patients | US | To examine the application of CEUS for automated FLL diagnosis using DNN. | DNN | This deep learning-based method provides comparable or better results, for an increased number of FLL types. |
Zhou et al., 2020 [18] | China | Frontiers in Oncology | 435 patients (616 liver lesions; 462 training set, 154 testing set) | CT | To propose a framework based on hierarchical CNNs for automatic detection and classification FLLs in multi-phasic CT. | Hierarchical CNNs | Overall, this preliminary study demonstrates that the proposed multi-modality and multi-scale CNN structure can locate and classify FLLs accurately in a limited dataset and would help inexperienced physicians to reach a diagnosis in clinical practice. |
Kim et al., 2020 [19] | South Korea | Scientific Reports | 549 patients, and external validation data set (54 patients) | MRI | To develop a fully automated deep learning model to detect HCC using hepatobiliary phase MR images and evaluate its performance in detecting HCC on liver MRI compared to human readers | Fine-tuned CNN | The optimised CNN architecture achieved 94% sensitivity, 99% specificity, and 0.97 area under curve (AUC) for HCC cases in the test dataset and achieved 87% sensitivity and 93% specificity and an AUC of 0.90 for external validation datasets. |
Huang et al., 2020 [20] | China | IEEE journal of biomedical and health informatics | Data set 1: 155 patients with FNH and 49 patients with atypical HCC Data set 2: 102 patients with FNH and only 36 patients with atypical HCC | US | To propose a novel liver tumour CAD approach extracting spatial-temporal semantics for atypical HCC. | SVM | The average accuracy reaches 94.40%, recall rate 94.76%, F1-score value 94.62%, specificity 93.62% and sensitivity 94.76%. |
Shi et al., 2020 [21] | NA | Abdominal Radiology | 342 patients with 449 untreated lesions (194 HCC group; 255 non-HCC group) | CT | To evaluate whether a three-phase dynamic contrast-enhanced CT protocol, when combined with a deep learning model, has similar accuracy in differentiating HCC from other FLLs) compared with a four-phase protocol. | CDNs | When combined with a CDN, a three-phase CT protocol without pre-contrast showed similar diagnostic accuracy as a four-phase protocol in differentiating HCC from other FLLs, suggesting that the multiphase CT protocol for HCC diagnosis might be optimised by removing the pre-contrast phase to reduce radiation dose. |
Zhen et al., 2020 [22] | China | Frontiers in Oncology | 1210 patients (31,608 images), and external validated cohort of 201 patients (6816 images) | MRI | To develop a DLS to classify liver tumours. | CNN | DLS that integrated these models could be used as an accurate and timesaving assisted-diagnostic strategy for liver tumours in clinical settings, even in the absence of contrast agents. DLS therefore has the potential to avoid contrast-related side effects and reduce economic costs associated with current standard MRI inspection practices for liver tumour patients. |
Krishan et al., 2020 [23] | India | Proceedings of the Institution of Mechanical Engineers, Part H: Journal of Engineering in Medicine | 794 normal liver images and 844 abnormal liver types (483 MET, 361 HCC) | CT | To detect the presence of a tumour region in the liver and classify the different stages of the tumour from CT images. | R-part decision tree, AdaBoost, RF, k-SVM, GLM, and NN. A multi-level ensemble model is also developed. | The accuracy achieved for different classifiers varies between 98.39% and 100% for tumour identification and between 76.38% and 87.01% for tumour classification. The multi-level ensemble model achieved high accuracy in both the detection and classification of different tumours. |
Brehar et al., 2020 [24] | Romania | Sensors | 268 patients | US | To compare deep-learning and conventional machine-learning methods for the automatic recognition of the HCC areas from US images | CNNs | The achieved results show that the deep-learning approach overcomes classical machine-learning solutions, by providing a higher classification performance. |
Hamm et al., 2019 [25] | NA | European radiology | 296 patients; 334 imaging studies; 494 hepatic lesions divided into training (434) and test sets (60) | MRI | To develop and validate a proof-of-concept CNN–based DLS that classifies common hepatic lesions on multi-phasic MRI. | CNN | This preliminary deep learning study demonstrated feasibility for classifying lesions with typical imaging features from six common hepatic lesion types. |
Das et al., 2019 [26] | India | Pattern Recognition and Image Analysis | 123 real-time images (63 HCC, and 60 metastasis carcinoma) | CT | To present an automatic approach that integrates the adaptive thresholding and spatial fuzzy clustering approach for detection of cancer region in CT scan images of liver. | Multilayer perceptron and C4.5 decision tree classifiers | This result proves that the spatial fuzzy c-means-based segmentation with C4.5 decision tree classifier is an effective approach for automatic recognition of the liver cancer. |
Trivizakis et al., 2019 [27] | Greece | IEEE Journal of Biomedical and Health Informatics | 130 images for the training and validation of the network | MRI | Propose and evaluate a novel 3D CNN designed for tissue classification in medical imaging and applied for discriminating between primary and metastatic liver tumours from diffusion weighted MRI data. | 3D CNN | The proposed 3D CNN architecture can bring significant benefit in diffusion weighted MRI liver discrimination and potentially, in numerous other tissue classification problems based on tomographic data, especially in size-limited, disease-specific clinical datasets. |
Kutlu et al., 2019 [28] | Turkey | Sensors | 56 images benign and 56 malignant liver tumours | CT | A new liver and brain tumour classification method is proposed by using the power of CNN in feature extraction, the power of DWT in signal processing, and the power of LSTM in signal classification. | CNN in feature extraction, DWT in signal processing, and LSTM in signal classification | The proposed method has a satisfactory accuracy rate at the liver tumour and brain tumour classifying. |
Nayak et al., 2019 [29] | India | International Journal of Computer Assisted Radiology and Surgery | 40 patients (healthy 14, cirrhosis 12, and cirrhosis with HCC 14) | CT | To proposes a CAD system for detecting cirrhosis and HCC in a very efficient and less time-consuming approach. | SVM | The proposed CAD system showed promising results and can be used as effective screening tool in medical image analysis. |
Schmauch et al., 2019 [30] | France | Diagnostic and Interventional Imaging | 544 patients (367 training set, 177 test set) | US | To create an algorithm that simultaneously detects and characterises (benign vs. malignant) FLL using deep learning. | ANN | This method could prove to be highly relevant for medical imaging once validated on a larger independent cohort. |
Jansen et al., 2019 [31] | The Netherlands | PLoS ONE | 95 patients (213 images) | MRI | Additional MR sequences and risk factors are used for automatic classification. | Randomised trees classifier | The proposed classification can differentiate five common types of lesions and is a step forward to a clinically useful aid. |
Das et al., 2019 [32] | India | Cognitive Systems Research | 75 patients (225 images) | CT | To introduce a new automated technique based on watershed–Gaussian segmentation approach. | DNN | The developed system is ready to be tested with huge database and can aid the radiologist in detecting the liver cancer using CT images. |
Mokrane et al., 2019 [33] | France | European Radiology | 178 patients (142 training set, 36 validations set) | CT | To enhance clinician’s decision-making by diagnosing HCC in cirrhotic patients with indeterminate liver nodules using quantitative imaging features extracted from triphasic CT scans. | KNN, SVM, and RF | A proof of concept that machine-learning-based radiomics signature using change in quantitative CT features across the arterial and portal venous phases can allow a non-invasive accurate diagnosis of HCCs in cirrhotic patients with indeterminate nodules. |
Lakshmipriya et al., 2019 [34] | India | Journal of Biomedical and Health Informatics | 634 images (440 images training set, 194 images validation set) | CT | An ensemble FCNet classifier is proposed to classify hepatic lesions from the deep features extracted using GoogleNetLReLU transfer learning approach. | CNN | Results demonstrate the efficacy of the proposed classifier design in achieving better classification accuracy. |
Acharya et al., 2018 [35] | Malaysia | Computers in biology and medicine | 101 patients with 463 images | US | This study initiates a CAD system to aid radiologists in an objective and more reliable interpretation of ultrasound images of liver lesions. | Radon transform and bi-directional empirical mode decomposition to extract features from the focal liver lesions. | The accuracy, sensitivity, and specificity of lesion classification were 92.95%, 90.80%, and 97.44%, respectively. |
Ta et al., 2018 [36] | USA | Radiology | 106 images (54 malignant, 51 benign, and one indeterminate FLL) | US | To assess the performance of CAD systems and to determine the dominant US features when classifying benign versus malignant FLLs by using contrast material–enhanced US cine clips. | ANN and SVM | CAD systems classified benign and malignant FLLs with an accuracy like that of an expert reader. CAD improved the accuracy of both readers. Time-based features of TIC were more discriminating than intensity-based features. |
Bharti et al., 2018 [37] | India | Ultrasonic Imaging | 94 patients (189 images) | US | To deal with this difficult visualisation problem, a method has been developed for classifying four liver stages, that is, normal, chronic, cirrhosis, and HCC evolved over cirrhosis. | KNN, SVM, rotation forest, CNNs | The experimental results strongly suggest that the proposed ensemble classifier model is beneficial for differentiating liver stages based on US images. |
Yasaka et al., 2018 [38] | Japan | Radiology | 560 patients (460 patients training set with 55,536 images, 100 patients validation set with 100 images) | CT | To investigate diagnostic performance by using a deep learning method with a CNN for the differentiation of liver masses at dynamic contrast agent-enhanced CT. | CNN | Deep learning with CNN showed high diagnostic performance in differentiation of liver masses at dynamic CT. |
Hassan et al., 2017 [39] | Egypt | Arabian Journal for Science and Engineering | 110 patients (110 images) | US | A new classification framework is introduced for diagnosing focal liver diseases based on deep learning architecture. | Stacked Sparse Autoencoder | Our proposed method presented overall accuracy of 97.2% compared with multi-SVM, KNN, and Naïve Bayes. |
Guo et al., 2017 [40] | China | Clinical Hemorheology and Microcirculation | 93 patients | US | To propose a novel two-stage multi-view learning framework for the CEUS based CAD for liver tumours, which adopted only three typical CEUS images selected from the arterial phase, portal venous phase and late phase. | Deep canonical correlation analysis and multiple kernel learning | The experimental results indicate that the proposed achieves best performance for discriminating benign liver tumours from malignant liver cancers. |
Kondo et al., 2017 [41] | Japan | Transactions on Medical Imaging | 98 patients | US | To propose an automatic classification method based on machine learning in CEUS of FLLs using the contrast agent Sonazoid. | SVM | The results indicated that combining the features from the arterial, portal, and post-vascular phases was important for classification methods based on machine learning for Sonazoid CEUS. |
Gatos et al., 2015 [42] | NA | Medical physics | 52 patients; (30 benign and 22 malignant) | US | Detect and classify FLLs from CEUS imaging by means of an automated quantification algorithm. | SVMs | The proposed quantification system that employs FLLs detection and classification algorithms may be of value to physicians as a second opinion tool for avoiding unnecessary invasive procedures. |
Virmani et al., 2014 [43] | India | Journal of Digital Imaging | 108 images | US | An NNE-based CAD system to assist radiologists in differential diagnosis between FLLs. | NNE | The promising results obtained by the proposed system indicate its usefulness to assist radiologists in differential diagnosis of FLLs. |
Wu et al., 2014 [44] | China | Optik | 22 patients | US | To propose a diagnostic system for liver disease classification based on CEUS imaging. | DNN | Quantitative comparisons demonstrate that the proposed method outperforms the compared classification methods in accuracy, sensitivity, and specificity |
Virmani et al., 2013 [45] | India | Defence Science Journal | 108 images comprising of 21 NOR images, 12 Cyst, 15 HEM, 28 HCC, and 32 MET | US | To investigate the contribution made by texture of regions inside and outside of the lesions in FLLs. | SVM | The proposed PCA-SVM based CAD system yielded classification accuracy of 87.2% with the individual class accuracy of 85%, 96%, 90%, 87.5%, and 82.2% for NOR, Cyst, HEM, HCC, and MET cases, respectively. The accuracy for typical, atypical, small HCC and large HCC cases is 87.5%, 86.8%, 88.8%, and 87%, respectively. |
Streba et al., 2012 [46] | Romania | World Journal of Gastroenterology | 224 patients | US | To study the role of time-intensity curve analysis parameters in a complex system of neural networks designed to classify liver tumours. | ANN | Neural network analysis of CEUS-obtained time-intensity curves seem a promising field of development for future techniques, providing fast and reliable diagnostic aid for the clinician. |
Mittal et al., 2011 [47] | India | Computerized Medical Imaging and Graphics | 88 patients with 111 images comprising 16 normal liver, 17 Cyst, 15 HCC, 18 HEM and 45 MET | US | It proposes a CAD system to assist radiologists in identifying focal liver lesions in B-mode ultrasound images. | Two step neural network classifier | The classifier has given correct diagnosis of 90.3% (308/340) in the tested segmented regions-of-interest from typical cases and 77.5% (124/160) in tested segmented regions-of-interest from atypical cases. |
Sugimoto et al., 2010 [48] | Japan | World Journal of Radiology | 137 patients (74 HCCs, 33 liver metastases and 30 liver hemangiomas) | US | To introduce CAD aimed at differential Diagnosis of FLLs by use of CEUS. | ANNs | The classification accuracies were 84.8% for metastasis, 93.3% for hemangioma, and 98.6% for all HCCs. In addition, the classification accuracies for histologic differentiation types of HCCs were 65.2% for w-HCC, 41.7% for m-HCC, and 80.0% for p-HCC. |
Shiraishi et al., 2008 [49] | Japan | Medical Physics | 97 patients, (103 images; 26 metastases, 16 hemangiomas, and 61 HCCs) | US | To develop a CAD scheme for classifying focal liver lesions as liver metastasis, hemangioma, and three histologic differentiation types of HCC, by use of microflow imaging of CEUS. | ANNs | The classification accuracies for the 103 FLLs were 88.5% for metastasis, 93.8% for hemangioma, and 86.9% for all HCCs. In addition, the classification accuracies for histologic differentiation types of HCCs were 79.2% for w-HCC, 50.0% for m-HCC, and 77.8% for p-HCC. |
Stoitsis et al., 2006 [50] | Greece | Nuclear Instruments and Methods in Physics Research | 147 images (normal liver 76, hepatic cyst 19, hemangioma 28, HCC 24) | CT | To classify of four types of hepatic tissue: normal liver, hepatic cyst, hemangioma, and hepatocellular carcinoma, from CT images. | Combined use of texture features and classifiers | The achieved classification performance was 100%, 93.75%, and 90.63% in the training, validation, and testing set, respectively. |
Matake et al., 2006 [51] | NA | Academic radiology | 120 patients | CT | To apply an ANN for differential diagnosis of certain hepatic masses on CT images and evaluate the effect of ANN output on radiologist diagnostic performance. | ANN | The ANN can provide useful output as a second opinion to improve radiologist diagnostic performance in the differential diagnosis of hepatic masses seen on contrast-enhanced CT. |
Gletsos et al., 2003 [52] | Greece | IEEE transactions on information technology in biomedicine | 147 patients | CT | To present a CAD system for the classification of hepatic lesions from CT images. | Neural-Network Classifier | The suitability of co-occurrence texture features, the superiority of GAs for feature selection, compared to sequential search methods, and the high performance achieved by the NN classifiers in the testing images set have been demonstrated. |
Chen et al., 1998 [53] | Taiwan | IEEE Transactions on Biomedical Engineering | 30 patients | CT | To present a CT liver image diagnostic classification system which will automatically find, extract the CT liver boundary, and further classify liver diseases. | Modified probabilistic NN | The proposed system was evaluated by 30 liver cases and shown to be efficient and very effective. |
4. Discussion
Limitations and Strengths
5. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Conflicts of Interest
References
- Shapiro, S.C. Encyclopaedia of Artificial Intelligence, 2nd ed.; John Wiley & Sons: Nashville, TN, USA, 1992; Volume 1. [Google Scholar]
- Calderaro, J.; Seraphin, T.P.; Luedde, T.; Simon, T.G. Artificial intelligence for the prevention and clinical management of hepatocellular carcinoma. J. Hepatol. 2022, 76, 1348–1361. [Google Scholar] [CrossRef] [PubMed]
- Sung, H.; Ferlay, J.; Siegel, R.L.; Laversanne, M.; Soerjomataram, I.; Jemal, A.; Bray, F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J. Clin. 2021, 71, 209–249. [Google Scholar] [CrossRef] [PubMed]
- Laino, M.E.; Viganò, L.; Ammirabile, A.; Lofino, L.; Generali, E.; Francone, M.; Lleo, A.; Saba, L.; Savevski, V. The Added Value of Artificial Intelligence to LI-RADS Categorization: A Systematic Review. Eur. J. Radiol. 2022, 150, 110251. Available online: https://www.ejradiology.com/article/S0720-048X(22)00101-2/fulltext (accessed on 5 July 2022). [CrossRef] [PubMed]
- Morland, D.; Triumbari, E.K.A.; Boldrini, L.; Gatta, R.; Pizzuto, D.; Annunziata, S. Radiomics in Oncological PET Imaging: A Systematic Review—Part 1, Supradiaphragmatic Cancers. Diagnostics 2022, 12, 1329. [Google Scholar] [CrossRef]
- The Jamovi Project. Jamovi. (Version 2.2) [Computer Software]. 2021. Available online: https://www.jamovi.org (accessed on 1 August 2022).
- R Core Team. R: A Language and Environment for Statistical Computing. (Version 4.0) [Computer Software]. 2021. Available online: https://cran.r-project.org (accessed on 1 August 2022).
- Ziegelmayer, S.; Reischl, S.; Harder, F.; Makowski, M.; Braren, R.; Gawlitza, J. Feature Robustness and Diagnostic Capabilities of Convolutional Neural Networks against Radiomics Features in Computed Tomography Imaging. Investig. Radiol. 2022, 57, 171–177. [Google Scholar] [CrossRef]
- Xu, X.; Mao, Y.; Tang, Y.; Liu, Y.; Xue, C.; Yue, Q.; Liu, Y.; Wang, J.; Yin, Y. Classification of Hepatocellular Carcinoma and Intrahepatic Cholangiocarcinoma Based on Radiomic Analysis. Comput. Math. Methods Med. 2022, 2022, 5334095. [Google Scholar] [CrossRef]
- Turco, S.; Tiyarattanachai, T.; Ebrahimkheil, K.; Eisenbrey, J.; Kamaya, A.; Mischi, M.; Lyshchik, A.; Kaffas, A.E. Interpretable Machine Learning for Characterization of Focal Liver Lesions by Contrast-Enhanced Ultrasound. IEEE Trans. Ultrason. Ferroelectr. Freq. Control 2022, 69, 1670–1681. [Google Scholar] [CrossRef]
- Sato, M.; Kobayashi, T.; Soroida, Y.; Tanaka, T.; Nakatsuka, T.; Nakagawa, H.; Nakamura, A.; Kurihara, M.; Endo, M.; Hikita, H.; et al. Development of novel deep multimodal representation learning-based model for the differentiation of liver tumors on B-mode ultrasound images. J. Gastroenterol. Hepatol. 2022, 37, 678–684. [Google Scholar] [CrossRef]
- Rela, M.; Rao, S.N.; Patil, R.R. Performance Analysis of Liver Tumor Classification Using Machine Learning Algorithms. IJATEE 2022, 9, 143. Available online: https://www.accentsjournals.org/paperInfo.php?journalPaperId=1393 (accessed on 20 July 2022).
- Zheng, R.; Wang, L.; Wang, C.; Yu, X.; Chen, W.; Li, Y.; Li, W.; Yan, F.; Wang, H.; Li, R. Feasibility of automatic detection of small hepatocellular carcinoma (≤2 cm) in cirrhotic liver based on pattern matching and deep learning. Phys. Med. Biol. 2021, 66, 8. [Google Scholar] [CrossRef]
- Yang, C.J.; Wang, C.K.; Fang, Y.H.D.; Wang, J.Y.; Su, F.C.; Tsai, H.M.; Lin, Y.J.; Tsai, H.W.; Yee, L.R. Clinical application of mask region-based convolutional neural network for the automatic detection and segmentation of abnormal liver density based on hepatocellular carcinoma computed tomography datasets. PLoS ONE 2021, 16, e0255605. [Google Scholar] [CrossRef] [PubMed]
- Stollmayer, R.; Budai, B.K.; Tóth, A.; Kalina, I.; Hartmann, E.; Szoldán, P.; Bérczi, V.; Maurovich-Horvat, P.; Kaposi, P.N. Diagnosis of focal liver lesions with deep learning-based multi-channel analysis of hepatocyte-specific contrast-enhanced magnetic resonance imaging. World J. Gastroenterol. 2021, 27, 5978–5988. [Google Scholar] [CrossRef]
- Kim, D.W.; Lee, G.; Kim, S.Y.; Ahn, G.; Lee, J.G.; Lee, S.S.; Kim, K.W.; Park, S.H.; Lee, Y.J.; Kim, N. Deep learning-based algorithm to detect primary hepatic malignancy in multiphase CT of patients at high risk for HCC. Eur. Radiol. 2021, 31, 7047–7057. [Google Scholar] [CrossRef] [PubMed]
- Căleanu, C.D.; Sîrbu, C.L.; Simion, G. Deep Neural Architectures for Contrast Enhanced Ultrasound (CEUS) Focal Liver Lesions Automated Diagnosis. Sensors 2021, 21, 4126. [Google Scholar] [CrossRef] [PubMed]
- Zhou, J.; Wang, W.; Lei, B.; Ge, W.; Huang, Y.; Zhang, L.; Yan, Y.; Zhou, D.; Ding, Y.; Wu, J.; et al. Automatic Detection and Classification of Focal Liver Lesions Based on Deep Convolutional Neural Networks: A Preliminary Study. Front. Oncol. 2020, 10, 581210. [Google Scholar] [CrossRef]
- Kim, J.; Min, J.H.; Kim, S.K.; Shin, S.Y.; Lee, M.W. Detection of Hepatocellular Carcinoma in Contrast-Enhanced Magnetic Resonance Imaging Using Deep Learning Classifier: A Multi-Center Retrospective Study. Sci. Rep. 2020, 10, 9458. [Google Scholar] [CrossRef]
- Huang, Q.; Pan, F.; Li, W.; Yuan, F.; Hu, H.; Huang, J.; Yu, J.; Wang, W. Differential Diagnosis of Atypical Hepatocellular Carcinoma in Contrast-Enhanced Ultrasound Using Spatio-Temporal Diagnostic Semantics. IEEE J. Biomed. Health Inform. 2020, 24, 2860–2869. [Google Scholar] [CrossRef]
- Shi, W.; Kuang, S.; Cao, S.; Hu, B.; Xie, S.; Chen, S.; Chen, Y.; Gao, D.; Chen, S.; Zhu, Y.; et al. Deep learning assisted differentiation of hepatocellular carcinoma from focal liver lesions: Choice of four-phase and three-phase CT imaging protocol. Abdom. Radiol. 2020, 45, 2688–2697. [Google Scholar] [CrossRef]
- Zhen, S.H.; Cheng, M.; Tao, Y.B.; Wang, Y.F.; Juengpanich, S.; Jiang, Z.Y.; Jiang, Y.K.; Yan, Y.Y.; Lu, W.; Lue, J.M.; et al. Deep Learning for Accurate Diagnosis of Liver Tumor Based on Magnetic Resonance Imaging and Clinical Data. Front. Oncol. 2020, 10, 680. [Google Scholar] [CrossRef]
- Ensembled Liver Cancer Detection and Classification Using CT Images—Abhay Krishan, Deepti Mittal. 2021. Available online: https://journals.sagepub.com/doi/abs/10.1177/0954411920971888?journalCode=pihb (accessed on 20 July 2022).
- Brehar, R.; Mitrea, D.A.; Vancea, F.; Marita, T.; Nedevschi, S.; Lupsor-Platon, M.; Rotaru, M.; Badea, R.I. Comparison of Deep-Learning and Conventional Machine-Learning Methods for the Automatic Recognition of the Hepatocellular Carcinoma Areas from Ultrasound Images. Sensors 2020, 20, 3085. [Google Scholar] [CrossRef]
- Hamm, C.A.; Wang, C.J.; Savic, L.J.; Ferrante, M.; Schobert, I.; Schlachter, T.; Lin, M.; Duncan, J.S.; Weinreb, J.C.; Chapiro, J.; et al. Deep learning for liver tumor diagnosis part I: Development of a convolutional neural network classifier for multi-phasic MRI. Eur. Radiol. 2019, 29, 3338–3347. [Google Scholar] [CrossRef] [PubMed]
- Das, A.; Das, P.; Panda, S.S.; Sabut, S. Detection of Liver Cancer Using Modified Fuzzy Clustering and Decision Tree Classifier in CT Images. Pattern Recognit. Image Anal. 2019, 29, 201–211. [Google Scholar] [CrossRef]
- Trivizakis, E.; Manikis, G.C.; Nikiforaki, K.; Drevelegas, K.; Constantinides, M.; Drevelegas, A.; Constantinides, M.; Drevelegas, A.; Marias, K. Extending 2-D Convolutional Neural Networks to 3-D for Advancing Deep Learning Cancer Classification with Application to MRI Liver Tumor Differentiation. IEEE J. Biomed. Health Inform. 2019, 23, 923–930. [Google Scholar] [CrossRef] [PubMed]
- Kutlu, H.; Avcı, E. A Novel Method for Classifying Liver and Brain Tumors Using Convolutional Neural Networks, Discrete Wavelet Transform and Long Short-Term Memory Networks. Sensors 2019, 19, 1992. [Google Scholar] [CrossRef] [Green Version]
- Nayak, A.; Baidya Kayal, E.; Arya, M.; Culli, J.; Krishan, S.; Agarwal, S.; Mehndiratta, A. Computer-aided diagnosis of cirrhosis and hepatocellular carcinoma using multi-phase abdomen CT. Int. J. Comput. Assist. Radiol. Surg. 2019, 14, 1341–1352. [Google Scholar] [CrossRef]
- Schmauch, B.; Herent, P.; Jehanno, P.; Dehaene, O.; Saillard, C.; Aubé, C.; Luciani, A.; Lassau, N.; Jégou, S. Diagnosis of focal liver lesions from ultrasound using deep learning. Diagn. Interv. Imaging 2019, 100, 227–233. [Google Scholar] [CrossRef]
- Automatic Classification of Focal Liver Lesions Based on MRI and Risk Factors. Available online: https://pubmed.ncbi.nlm.nih.gov/31095624/ (accessed on 20 July 2022).
- Das, A.; Acharya, U.R.; Panda, S.S.; Sabut, S. Deep learning based liver cancer detection using watershed transform and Gaussian mixture model techniques. Cogn. Syst. Res. 2019, 54, 165–175. [Google Scholar] [CrossRef]
- Mokrane, F.Z.; Lu, L.; Vavasseur, A.; Otal, P.; Peron, J.M.; Luk, L.; Yang, H.; Ammari, S.; Saenger, Y.; Rousseau, H. Radiomics machine-learning signature for diagnosis of hepatocellular carcinoma in cirrhotic patients with indeterminate liver nodules. Eur. Radiol. 2020, 30, 558–570. [Google Scholar] [CrossRef]
- Balagourouchetty, L.; Pragatheeswaran, J.K.; Pottakkat, B.; Ramkumar, G. GoogLeNet-Based Ensemble FCNet Classifier for Focal Liver Lesion Diagnosis. IEEE J. Biomed. Health Inform. 2020, 24, 1686–1694. [Google Scholar] [CrossRef]
- Acharya, U.R.; Koh, J.E.W.; Hagiwara, Y.; Tan, J.H.; Gertych, A.; Vijayananthan, A.; Yaakup, N.A.; Abdullah, B.J.J.; Bin Mohd Fabell, M.K.; Yeong, C.H. Automated diagnosis of focal liver lesions using bidirectional empirical mode decomposition features. Comput. Biol. Med. 2018, 94, 11–18. [Google Scholar] [CrossRef]
- Ta, C.N.; Kono, Y.; Eghtedari, M.; Oh, Y.T.; Robbin, M.L.; Barr, R.G.; Kummel, A.C.; Mattrey, R.F. Focal Liver Lesions: Computer-aided Diagnosis by Using Contrast-enhanced US Cine Recordings. Radiology 2018, 286, 1062–1071. [Google Scholar] [CrossRef] [PubMed]
- Bharti, P.; Mittal, D.; Ananthasivan, R. Preliminary Study of Chronic Liver Classification on Ultrasound Images Using an Ensemble Model. Ultrason. Imaging 2018, 40, 357–379. [Google Scholar] [CrossRef] [PubMed]
- Deep Learning with Convolutional Neural Network for Differentiation of Liver Masses at Dynamic Contrast-Enhanced CT: A Preliminary Study. Available online: https://pubmed.ncbi.nlm.nih.gov/29059036/ (accessed on 20 July 2022).
- Hassan, T.M.; Elmogy, M.; Sallam, E.S. Diagnosis of Focal Liver Diseases Based on Deep Learning Technique for Ultrasound Images. Arab. J. Sci. Eng. 2017, 42, 3127–3140. [Google Scholar] [CrossRef]
- Guo, L.H.; Wang, D.; Qian, Y.Y.; Zheng, X.; Zhao, C.K.; Li, X.L.; Bo, X.W.; Yue, W.W.; Zhang, Q.; Shi, J.; et al. A two-stage multi-view learning framework based computer-aided diagnosis of liver tumors with contrast enhanced ultrasound images. Clin. Hemorheol. Microcirc. 2018, 69, 343–354. [Google Scholar] [CrossRef]
- Kondo, S.; Takagi, K.; Nishida, M.; Iwai, T.; Kudo, Y.; Ogawa, K.; Kamiyama, T.; Shibuya, H.; Kahata, K.; Shimizu, C. Computer-Aided Diagnosis of Focal Liver Lesions Using Contrast-Enhanced Ultrasonography with Perflubutane Microbubbles. IEEE Trans. Med. Imaging 2017, 36, 1427–1437. [Google Scholar] [CrossRef]
- Gatos, I.; Tsantis, S.; Spiliopoulos, S.; Skouroliakou, A.; Theotokas, I.; Zoumpoulis, P.; Hazle, J.D.; Kagadis, G.C. A new automated quantification algorithm for the detection and evaluation of focal liver lesions with contrast-enhanced ultrasound. Med. Phys. 2015, 42, 3948–3959. [Google Scholar] [CrossRef]
- Virmani, J.; Kumar, V.; Kalra, N.; Khandelwal, N. Neural network ensemble based CAD system for focal liver lesions from B-mode ultrasound. J. Digit. Imaging 2014, 27, 520–537. [Google Scholar] [CrossRef] [Green Version]
- Wu, K.; Chen, X.; Ding, M. Deep learning based classification of focal liver lesions with contrast-enhanced ultrasound. Optik 2014, 125, 4057–4063. [Google Scholar] [CrossRef]
- Virmani, J.; Kumar, V.; Kalra, N.; Khandelwal, N. Characterization of Primary and Secondary Malignant Liver Lesions from B-Mode Ultrasound. J. Digit. Imaging 2013, 26, 1058–1070. [Google Scholar] [CrossRef]
- Streba, C.T.; Ionescu, M.; Gheonea, D.I.; Sandulescu, L.; Ciurea, T.; Saftoiu, A.; Vere, C.C.; Rogoveanu, I. Contrast-enhanced ultrasonography parameters in neural network diagnosis of liver tumors. World J. Gastroenterol. 2012, 18, 4427–4434. [Google Scholar] [CrossRef]
- Mittal, D.; Kumar, V.; Saxena, S.C.; Khandelwal, N.; Kalra, N. Neural network based focal liver lesion diagnosis using ultrasound images. Comput. Med. Imaging Graph. 2011, 35, 315–323. [Google Scholar] [CrossRef] [PubMed]
- Sugimoto, K.; Shiraishi, J.; Moriyasu, F.; Doi, K. Computer-aided diagnosis for contrast-enhanced ultrasound in the liver. World J. Radiol. 2010, 2, 215–223. [Google Scholar] [CrossRef]
- Shiraishi, J.; Sugimoto, K.; Moriyasu, F.; Kamiyama, N.; Doi, K. Computer-aided diagnosis for the classification of focal liver lesions by use of contrast-enhanced ultrasonography. Med. Phys. 2008, 35, 1734–1746. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Stoitsis, J.; Valavanis, I.; Mougiakakou, S.G.; Golemati, S.; Nikita, A.; Nikita, K.S. Computer aided diagnosis based on medical image processing and artificial intelligence methods. Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrometers Detect. Assoc. Equip. 2006, 569, 591–595. [Google Scholar] [CrossRef]
- Usefulness of Artificial Neural Network for Differential Diagnosis of Hepatic Masses on CT Images—ScienceDirect. Available online: https://www.sciencedirect.com/science/article/abs/pii/S1076633206002613 (accessed on 20 July 2022).
- Gletsos, M.; Mougiakakou, S.G.; Matsopoulos, G.K.; Nikita, K.S.; Nikita, A.S.; Kelekis, D. A computer-aided diagnostic system to characterize CT focal liver lesions: Design and optimization of a neural network classifier. IEEE Trans. Inf. Technol. Biomed. 2003, 7, 153–162. [Google Scholar] [CrossRef]
- Chen, E.L.; Chung, P.C.; Chen, C.L.; Tsai, H.M.; Chang, C.I. An automatic diagnostic system for CT liver image classification. IEEE Trans. Biomed. Eng. 1998, 45, 783–794. [Google Scholar] [CrossRef] [Green Version]
- Short Overview of the Human Brain Project. Available online: https://www.humanbrainproject.eu/en/about/overview/ (accessed on 2 August 2022).
- Data-Efficient and Weakly Supervised Computational Pathology on Whole-Slide Images. Available online: https://pubmed.ncbi.nlm.nih.gov/33649564/ (accessed on 5 July 2022).
- Wu, B.; Khong, P.L.; Chan, T. Automatic detection and classification of nasopharyngeal carcinoma on PET/CT with support vector machine. Int. J. Comput. Assist. Radiol. Surg. 2012, 7, 635–646. [Google Scholar] [CrossRef] [Green Version]
- Kirienko, M.; Cozzi, L.; Rossi, A.; Voulaz, E.; Antunovic, L.; Fogliata, A.; Chiti, A.; Sollini, M. Ability of FDG PET and CT radiomics features to differentiate between primary and metastatic lung lesions. Eur. J. Nucl. Med. Mol. Imaging 2018, 45, 1649–1660. [Google Scholar] [CrossRef]
- Gao, X.; Chu, C.; Li, Y.; Lu, P.; Wang, W.; Liu, W.; Yu, L. The method and efficacy of support vector machine classifiers based on texture features and multi-resolution histogram from (18)F-FDG PET-CT images for the evaluation of mediastinal lymph nodes in patients with lung cancer. Eur. J. Radiol. 2015, 84, 312–317. [Google Scholar] [CrossRef]
- Wu, J.; Aguilera, T.; Shultz, D.; Gudur, M.; Rubin, D.L.; Loo, B.W.; Diehn, M.; Li, R. Early-Stage Non-Small Cell Lung Cancer: Quantitative Imaging Characteristics of (18)F Fluorodeoxyglucose PET/CT Allow Prediction of Distant Metastasis. Radiology 2016, 281, 270–278. [Google Scholar] [CrossRef] [Green Version]
- Sacks, A.; Peller, P.J.; Surasi, D.S.; Chatburn, L.; Mercier, G.; Subramaniam, R.M. Value of PET/CT in the management of liver metastases, part 1. AJR Am. J. Roentgenol. 2011, 197, W256–W259. [Google Scholar] [CrossRef] [PubMed]
- Khan, M.A.; Combs, C.S.; Brunt, E.M.; Lowe, V.J.; Wolverson, M.K.; Solomon, H. Positron emission tomography scanning in the evaluation of hepatocellular carcinoma. J. Hepatol. 2000, 32, 792–797. [Google Scholar] [CrossRef]
- Delbeke, D.; Martin, W.H.; Sandler, M.P.; Chapman, W.C.; Wright, J.K.; Pinson, C.W. Evaluation of benign vs malignant hepatic lesions with positron emission tomography. Arch. Surg. 1998, 133, 510–515; discussion 515–516. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Ho, C.; Chen, S.; Yeung, D.W.C.; Cheng, T.K.C. Dual-tracer PET/CT imaging in evaluation of metastatic hepatocellular carcinoma. J. Nucl. Med. 2007, 48, 902–909. [Google Scholar] [CrossRef] [Green Version]
- Park, J.W.; Kim, J.H.; Kim, S.K.; Kang, K.W.; Park, K.W.; Choi, J.I.; Lee, W.J.; Kim, C.M.; Nam, B.H. A prospective evaluation of 18F-FDG and 11C-acetate PET/CT for detection of primary and metastatic hepatocellular carcinoma. J. Nucl. Med. 2008, 49, 1912–1921. [Google Scholar] [CrossRef]
Levels | Counts | % of Total | Cumulative % |
---|---|---|---|
1998 | 1 | 2.2% | 2.2% |
2003 | 1 | 2.2% | 4.3% |
2006 | 2 | 4.3% | 8.7% |
2008 | 1 | 2.2% | 10.9% |
2010 | 1 | 2.2% | 13.0% |
2011 | 1 | 2.2% | 15.2% |
2012 | 1 | 2.2% | 17.4% |
2013 | 1 | 2.2% | 19.6% |
2014 | 2 | 4.3% | 23.9% |
2015 | 1 | 2.2% | 26.1% |
2017 | 3 | 6.5% | 32.6% |
2018 | 4 | 8.7% | 41.3% |
2019 | 10 | 21.7% | 63.0% |
2020 | 7 | 15.2% | 78.3% |
2021 | 5 | 10.9% | 89.1% |
2022 | 5 | 10.9% | 100.0% |
Median | 2019 |
Levels | Counts | % of Total | Cumulative % |
---|---|---|---|
0 | 4 | 8.7% | 8.7% |
1 | 6 | 13.0% | 21.7% |
2 | 36 | 78.3% | 100.0% |
Mean: 1.70 | Median: 2.00 | SD: 0.628 |
Levels | Counts | % of Total | Cumulative % |
---|---|---|---|
0 | 20 | 43.5% | 43.5% |
1 | 24 | 52.2% | 95.7% |
2 | 2 | 4.3% | 100.0% |
Mean: 0.609 | Median: 1.00 | SD: 0.577 |
Levels | Counts | % of Total | Cumulative % |
---|---|---|---|
1 | 6 | 13.0% | 13.0% |
2 | 40 | 87.0% | 100.0% |
Mean: 1.87 | Median: 2.00 | SD: 0.341 |
Levels | Counts | % of Total | Cumulative % |
---|---|---|---|
1 | 1 | 2.2% | 2.2% |
2 | 2 | 4.3% | 6.5% |
3 | 7 | 15.2% | 21.7% |
4 | 16 | 34.8% | 56.5% |
5 | 18 | 39.1% | 95.7% |
6 | 2 | 4.3% | 100.0% |
Mean: 4.17 | Median: 4.00 | SD: 1.04 |
Year | χ2 Tests | Year | χ2 Tests | Year | χ2 Tests | Year | χ2 Tests | Year | χ2 Tests | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Total Score | 1998 | 2003 | 2006 | 2008 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2017 | 2018 | 2019 | 2020 | 2021 | 2022 | Total | Value | df | p |
1 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 111 | 75 | 0.004 |
2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 2 | |||
3 | 0 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 1 | 2 | 0 | 1 | 0 | 0 | 0 | 7 | |||
4 | 0 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 1 | 3 | 3 | 2 | 1 | 3 | 16 | |||
5 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 5 | 3 | 4 | 2 | 18 | |||
6 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 2 | |||
Total | 1 | 1 | 2 | 1 | 1 | 1 | 1 | 1 | 2 | 1 | 3 | 4 | 10 | 7 | 5 | 5 | 46 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Martinino, A.; Aloulou, M.; Chatterjee, S.; Scarano Pereira, J.P.; Singhal, S.; Patel, T.; Kirchgesner, T.P.-E.; Agnes, S.; Annunziata, S.; Treglia, G.; et al. Artificial Intelligence in the Diagnosis of Hepatocellular Carcinoma: A Systematic Review. J. Clin. Med. 2022, 11, 6368. https://doi.org/10.3390/jcm11216368
Martinino A, Aloulou M, Chatterjee S, Scarano Pereira JP, Singhal S, Patel T, Kirchgesner TP-E, Agnes S, Annunziata S, Treglia G, et al. Artificial Intelligence in the Diagnosis of Hepatocellular Carcinoma: A Systematic Review. Journal of Clinical Medicine. 2022; 11(21):6368. https://doi.org/10.3390/jcm11216368
Chicago/Turabian StyleMartinino, Alessandro, Mohammad Aloulou, Surobhi Chatterjee, Juan Pablo Scarano Pereira, Saurabh Singhal, Tapan Patel, Thomas Paul-Emile Kirchgesner, Salvatore Agnes, Salvatore Annunziata, Giorgio Treglia, and et al. 2022. "Artificial Intelligence in the Diagnosis of Hepatocellular Carcinoma: A Systematic Review" Journal of Clinical Medicine 11, no. 21: 6368. https://doi.org/10.3390/jcm11216368