Table A1.

Qualitative data distilled from each included review on imaging modalities, AI techniques, and the corresponding review highlights.

No.	References	Imaging Modalities	AI Techniques	Cited By *	Highlights
Traditional Review
1	[84]	Thermography	CNN	21	Summarize the recent progress in breast cancer detection based on optical imaging (i.e., thermography) CNNs Application of deep neural networks for breast thermogram classification Future works in development of datasets, feeding the segmented image, assigning good kernel, and construction of a lightweight CNN model to enhance the CNN performance
2	[85]	Thermography	ANN: RBFN, KNN, PNN, SVM, ResNet50, SeResNet50, V Net, Bayes Net, CNN, C-DCNN, VGG-16, Hybrid (ResNet-50 and V-Net), ResNet101, DenseNet, and InceptionV3	11	Comprehensive review on related progress with AI in optical imaging (i.e., thermography) Highlight the contributions and limitations of each study Identify research gap Identify bottlenecks of ANN in breast thermography
3	[86]	X-ray, CT, ultrasound, MRI, nuclear, and microscopy	Deep learning	38	Summarize different medical imaging technologies Highlight the need of efficient medical data management techniques Summarize the advent algorithms for disease classification and organ/tissue segmentation using AI methods
4	[87]	Mammogram, ultrasound, and MRI	Machine learning and deep learning	1	Summarize the AI application in mammogram, ultrasound, and MRI Discuss the challenges and road ahead of AI in medical imaging
5	[88]	Mammogram, ultrasound, MRI, FDG PET/CT	Augmented intelligence and machine learning	16	Comprehensive overview on augmented intelligence in different breast imaging modalities
6	[89]	Mammogram, ultrasound, MRI,	Machine learning, CADe, ANN, CNN, and NLP	36	Discuss the potential of AI in breast imaging Identify the limitations and bottlenecks of AI application
7	[90]	Mammogram and ultrasound	Machine learning	1	Summarize the essential elements in data engineering, typically preparation of data for machine learning application Comprehensive review on databases, perspective, data integrity, and characteristic of a good data in the context of machine learning application
8	[91]	Thermography and electrical impedance tomography	Machine learning and CADe	15	Summarize the background, technical characteristics of machine learning techniques and CADe in thermography and electrical impedance tomography Identify the methods to enhance global performance Highlight recommendations for 3D breast simulation, pre-processing techniques, as well as relevant biomedical modality that has the potential in tumor localization and tumor size prediction
9	[92]	Transillumination imaging, diffuse optical imaging, and near-infrared spectroscopy	Machine learning	13	Comprehensive review on applicability (i.e., diagnostic ability and practical deployment) of the new spectroscopy techniques in discrimination of normal and cancerous tissue
10	[93]	Mammogram, Tomosynthesis, ultrasound, DBCT, MRI, DWI, CT, NIR fluorescence, and SPECT	SVM, ANN, and robotics	6	Overview on past and present AI methods in different breast cancer diagnostic imaging modalities
11	[94]	PET and MRI	Fuzzy logic and neural network	26	Summarize recent progress, key strengths, and weakness of PET/MRI in cancer imaging
12	[9]	Mammogram, tomosynthesis, DCE-MRI, and ultrasound	CADe and CADx systems	449	Comprehensive state-of-the-art review on AI approaches in medical imaging for four tumor types: lung, brain, breast, and prostate Identify the common clinical problem in medical imaging
13	[95]	MRI	Analytical radiomic-based (human-engineered) and deep learning-based CADe	44	Discuss the goal of AI in the context of breast cancer imaging, typically in MRI Review on current progress, potential application, and limitations of AI in breast MRI
14	[96]	MRI	Machine learning and deep learning	4	Discuss the current progress and future perspective of AI in breast MRI Highlight the importance of developing quantitative imaging biomarkers for precision medicine Discuss potential of machine learning and deep learning in breast MRI Discuss the future challenge of deep learning in the context of breast MRI and recommend AI-augmented clinical decision techniques
15	[97]	MRI	3D printing, augmented reality. Radiomics, and machine learning	12	Discuss the role of breast MRI in detection of DCIS with the aid of AI approaches
16	[98]	Mammogram, ultrasound, and MRI	ANN, CNN, CADe, and GANs	0	Comprehensive review on insight, impact, and role of AI in breast imaging
17	[99]	Mammogram	Machine learning (supervised, unsupervised, reinforcement, and deep learning)	6	Highlight and discuss key terminology, concepts, and common AI models in breast imaging
18	[100]	Mammogram and DBT	Deep learning	89	Comprehensive review on current progress and head ahead of the AI-based clinical application in mammogram, DBT, and radiomics
19	[101]	Ultrasonography	CNN	9	Current progress and potential application of AI models in breast ultrasonography
20	[102]	Mammogram, ultrasound, and MRI	Deep learning and CADe	32	Comprehensive review on state-of-the-art deep learning methods in mammogram, ultrasound, and MRI Discuss the challenges and bottlenecks which may impinge the advancement and integration of AI in clinical imaging workflow
21	[103]	Mammogram	Machine learning and radiomics	19	Discuss the CEM technique, screening, and diagnostic uses, and future applications with AI and radiomics
22	[104]	Mammogram, sonography, MRI, and image-guided biopsy	Deep learning and radiomics	2	Recent progress and potential value of digital analysis in breast imaging
23	[105]	Mammogram, ultrasound, PET, and MRI	ANN, SVM, and radiomics	1	Discuss key concept, state-of-the-art studies, and potential clinical use cases in breast imaging
24	[106]	Ultrasound	Deep learning	5	Discuss recent progress of deep learning in ultrasound imaging Discuss the workflow enhancement, typically on view recognition, image quality assessment, scanning guide, and quantification Recommend future perspective of deep learning in imaging workflow
25	[107]	Mammogram and ultrasound	Eye tracking tool and CADe	10	Highlight the importance of visual search in breast imaging Identify the underlying reasons for diagnostic error in breast imaging
26	[108]	Mammogram	CADe, CADx, machine learning, deep learning, and CNN	48	Review of recent progress of AI approaches in breast imaging Commentary on AI in medical diagnostic imaging
27	[109]	Mammogram and tomosynthesis	Deep learning	0	Synthesis recent progress of AI in mammography analyzing the risk in breast cancer Highlight the key priority, challenges, and future prospective of AI application in clinical workflow
28	[110]	Nuclear medicine	Deep learning and radiomics	1	Introduce the basic design of neural network Review of influential works, strengths, and limitations of deep learning approaches as compared to the conventional approaches
29	[111]	MRI, CT, PET, SPECT, ultrasound, tomosynthesis, and radiology	Neural network, deep learning, and machine learning	227	Discuss the basic terms of AI (e.g., machine and deep learning) and analyze the integration of AI in radiology
30	[112]	Mammogram, ultrasound, MRI, and tomosynthesis	ANN, CADe, CADx, CNN, deep learning, and machine learning	8	Overview on current application of deep learning approaches in breast radiology Identify and highlight the challenges and research gap of AI application in breast radiology
31	[113]	Mammogram, ultrasound, and MRI	Deep learning	5	Review of current knowledge and future prospective of AI-enhanced breast imaging modalities in clinical workflow
32	[114]	Mammogram, ultrasound, MRI	SNN, SDAE, DBN, and CNN	2	Review on state-of-the-art AI approaches in the last decade to detect breast cancer using different breast imaging modalities
33	[115]	Mammogram, ultrasound, MRI, and tomosynthesis	Deep learning and AI-CADe	79	Review the current limitations and future prospective of AI-CADe in breast imaging
34	[116]	Tomosynthesis, CT, and FDG PET/CT	ANN, DNN, SVM	84	Review the recent progress and strength of AI applications in cancer diagnostic and prognostic Identify how AI technology can improve cancer diagnostic and prognostic from the clinical perspective
35	[117]	Ultrasound	Machine learning and deep learning	8	General overview on the context of AI, machine learning, and deep learning technologies Review of current progress in deep learning technologies and highlight future perspective, challenges, and research gap of biomedical AI systems in ultrasound
36	[118]	Mammogram, tomosynthesis, ultrasonography, and MRI	CADe, radiomics, IoT, and machine learning tools	17	Critical review on advent methods for screening, diagnostic, staging, grading, molecular, and genetic biomarkers Highlight the interdisciplinary key terms of such methods for scientists and physicians
37	[119]	Mammogram and CT	Deep learning, CADe	6	Highlight major areas in oncology imaging with great potential of AI application Identify potential bottlenecks of AI application in oncology imaging
38	[120]	Mammogram, ultrasound, PET, CT, and MRI	CADe, ANN	127	Outline the progress of AI and CADe in general clinical imaging
39	[121]	Ultrasound	CADx	20	Highlight recent progress in CADx in breast ultrasound Recommend future direction of CADx in breast ultrasound
40	[122]	Mammogram, ultrasound, MRI, and thermography	Machine learning, deep learning, and CADx	42	Highlight new application of deep learning and machine learning approaches in detection and classification of breast cancer and outline the overview of such progress Review different breast imaging modalities and correlate these modalities with AI
41	[123]	Mammogram	Radiomics	0	Highlight the potential role of radiomics in mammography Identify the bottlenecks and research gap of such application
42	[124]	MRI and DCE-MRI	Radiogenomics	0	Review multidimensional mining algorithms in the context of MRI radiogenomics for CADe and CADx of breast tumors
43	[125]	Tomosynthesis, MRI, ultrasound, MBI	Machine learning and CADx	3	Review the current progress in imaging of high-density breast for diagnostic of breast lesions
44	[126]	Thermography	SVM, ANN, and CADx	42	Review of recent advances, recommendation, and road ahead for detection of breast cancer using infrared thermography
45	[127]	Mammogram	AI-CADe	62	Review of past, present, and future role of AI in medical imaging Highlight the challenges of AI application in imaging Highlight the role of AI-CADe with PACS
46	[128]	Mammogram and MRI	CNN	1	Highlight the fundamental in AI and its application in medical imaging analysis
47	[70]	General breast imaging	Machine learning, ANN, deep learning	11	Explore the evaluation and use of AI in breast imaging Explore the application of AI in breast imaging from the ethical, technical, and legal perspective
Purpose Specific Review
48	[129]	Mammogram, ultrasound, SWE, SWV, and sonoelastography	CADx, SVM, CNN, LASSO, and ridge regression	4	Highlight the functions of ultrasomics and its capability in disease diagnostic (precision medicine) in different organs Highlight the limitations, challenges, and opportunities in ultrasomics application
49	[130]	Mammogram, ultrasound, and MRI	Deep learning and radiogenomics	13	Highlight the functions and integration process of radiogenomics Remark the strategies and relevant algorithms in such application
50	[131]	Mammogram, ultrasound, and tomosynthesis	Machine learning and deep learning	54	Highlight the functions and role of AI in early detection of breast cancer Highlight the readiness of AI in breast cancer imaging
51	[132]	EIT	PSO, ANN, GA, and other machine learning algorithms	44	Highlight the functions and roles of EIT and its application with AI
52	[133]	DOT	Deep learning	3	Highlight the roles, background, and relevant tools of AI in DOT
53	[134]	Ultrasound	Deep learning	2	Highlight the roles of AI in cancer, dermatology, cardiology, respiratory problems, neurodegenerative disorders, and gastroenterology
54	[135]	PET/CT and PET/MRI	Radiomics	16	Highlight the solutions and challenges of radiomics application in hybrid PET/CT and PET/MRI
55	[136]	Mammogram	CADe and CADx	28	Highlight the roles of CADe and CADx in mammography
56	[137]	Mammogram, ultrasound, DBT, and MRI	Deep learning, CADe, and CADx	3	Highlight the functions of AI applications in medical tasks and reasons for AI to be classified as a medical device Review the functions of CADe and its sub-elements Outline the FDA process in approving the CADe software
57	[138]	Ultrasound	CADe and CNN	0	Highlight the variation of ultrasound for dense breast imaging amongst average risk women
58	[139]	Thermography	SVM, ANN, BN, CADe, CNN, and GA	0	Highlight the roles of AI in thermography for early detection in breast cancer
59	[140]	Thermography	ANN	13	Highlight the roles of infrared thermography in breast imaging associated with clinical evidence that favor and unfavored such application Outline the historical timeline of infrared thermography
Systematic Review
60	[141]	Mammogram, ultrasound, MRI, DBPET, DWI, PWI, CT, PET/CT, and PET/MRI	Radiomics, machine learning, and deep learning	5	Analyze and compare the conventional and advent breast imaging techniques Highlight the roles and future perspectives of advent breast imaging techniques in prediction of the response to NAC
61	[142]	Thermography	SVM, ANN, DNN, and RNN	81	Analyze and compare the studies in breast cancer detection using CADe and deep learning models
62	[143]	Mammogram, CT, and MRI	Machine learning, deep learning, and ANN,	1	Analyze the potential of AI in medical imaging Discuss the methodology, application, limitations, challenges, and future perspective in radiology
63	[144]	Mammogram, ultrasound, CT, and MRI	Deep CNN	2	Systematic review of current progress and state-of-the-art methods in intrafractional target motion management in breast cancer radiation therapy
64	[65]	Mammogram, DCE-MRI	Machine learning and deep learning	3	Systematic survey on 185 papers in cancer prediction and diagnostic using AI
65	[64]	Mammogram, ultrasound	CNN, ANN, DNN, MLP, SVM, DT, GA, KNN, NB, LR, LA, and GMM	12	Structured review on image processing, machine learning, and deep learning in breast imaging
Mixed Method Review
66	[145]	Ultrasound and tomosynthesis	DNN, RCNN, faster RCNN, deep CNN, and ReLU	4	Commentary on current progress of DNN in computational radiology in the context of breast imaging and diagnostic
67	[146]	Mammogram, tomosynthesis, ultrasound, tomography, and MRI	Multilayered DNN	12	Review of novel approaches in breast imaging aided with AI
68	[147]	Mammogram and MRI	Radiomics	22	Review of various medical imaging modalities in the context of early detection Highlight the applications of various medical imaging modalities, the respective benefits, and the application of radiomics in such modalities
69	[148]	MRI	Machine learning and transfer learning	0	Systematic survey on application of machine learning in assessment of ALNM using MRI
70	[149]	Mammogram, ultrasound, and MRI	ANN, SNN, CNN, and CADe	41	Systematic survey of breast cancer classification in different breast imaging modalities aided by DNN approaches
Qualitative Review
71	[150]	Mammogram, CT, and MRI	Machine learning	51	Qualitative survey of the impact of AI on radiology

* Data retrieved from the Scopus database as of 17 June 2022. AI: artificial intelligence; CNN: Convolutional Neural Networks; ANN: Artificial Neural Network; RBFN: Radial Basis Function Network; KNN: K-Nearest Neighbors; PNN: Probability Neural Network; SVM: Support Vector Machine; C-DCNN: DeConvolutional Neural Networks; CT: Computer tomography; MRI: magnetic resonance imaging; FDG: fluorodeoxyglucose; PET: Positron Emission Tomography; CADe: computer-aided detection; NLP: natural language processing; 3D: three dimensional; DBCT: Dedicated breast computed tomography; DWI: diffusion-weighted imaging, NIR: Near-Infrared; SPECT: Single-photon emission computed tomography; CADx: computer-aided diagnosis; DCE-MRI: dynamic contrast-enhanced magnetic resonance imaging; DCIS: ductal carcinoma in situ; GANs: generative adversarial networks; DBT: digital breast tomosynthesis; CEDM: Contrast-enhanced digital mammography; SNN: Shallow Neural Network; SDAE: Stacked Denoising Autoencoder; DBN: Deep Belief Network; DNN: deep neural networks; IoT: Internet of things; MBI: molecular breast imaging; PACS: picture archiving and communications system; SWE: shear wave elastography; SWV: shear wave viscosity; LASSO: least absolute shrinkage and selection operator; MRE: Magnetic resonance elastography; EIT: Electrical impedance tomography; PSO: particle swarm optimization; GA: genetic algorithm; DOT: Diffuse optical tomography; BN: Bayesian network; CE-MRI: contrast-enhanced magnetic resonance imaging; NAC: Neoadjuvant chemotherapy; DBPET: Dedicated breast positron emission tomography; PWI: perfusion weighted imaging; RNN: recurrent neural network; MLP: Multi-layer Perceptron; DT: decision tree; NB: naïve bayes; LR: Logistic Regression; LA: Linear discriminant analysis; GMM: Gaussian Mixture Modelling; RCNN: Region-Based Convolutional Neural Network; ReLU: rectified linear unit; ALNM: axillary lymph node metastasis.