TABLE 1.

Summary of ultrasomics studies in oncology.

Stydies	Study design	Cancer	No. of patients	Modality	Features	Feature classifier	Type of features	Statistical analysis	Endpoint	Result
Zhao et al. (37)	Retrospective Single center	Liver	177	BMUS/SWE/SWV	2560	SRT/SVM	GM/GEM/GEVM	Mann–Whitney U test	prognosis and diagnosis	AUC: 0.94 (benign/malignant) AUC: 0.97 (malignant subtyping) AUC: 0.97 (PD-1 prediction) AUC: 0.94 (Ki-67 prediction) AUC: 0.98 (MVI prediction)
Zhou et al. (20)	Retrospective Single center	Breast	205	SWE	4224	CNN	—	—	Diagnosis	Accuracy: 95.8% Sensitivity: 96.2% Specificity: 95.7%
Li et al. (21)	Retrospective Single center	Breast	178	BMUS/SWE/CEUS	1226	SVM	Intensity/Texture/Contourlet/Shape/Perfusion	Holdout test	Diagnosis	Accuracy: 84.12% Sensitivity: 92.86% Specificity:78.80% AUC: 0.919
Luo et al. (26)	Retrospective Single center	Breast	315	BMUS	1044	LASSO	Histogram/Texture/RLM/Form factor	Multivariate regression analysis	Diagnosis	AUC: 0.928
Lee et al. (30)	Retrospective Single center	Breast	901	BMUS	730	LASSO	Intensity/Texture/Wavelet	—	Diagnosis	AUC: 0.782
Zhang et al. (14)	Retrospective Single center	Breast	117	Sonoelastography	364	clusters derived	Shape/intensity/GLCM/contourlet	Clusters derived/SVM	Diagnosis	AUC: 0.97 Accuracy: 88.0% Sensitivity: 85.7% Specificity: 89.3%
Qiu et al. (31)	Retrospective Single center	Lymph node	256	BMUS	843	LASSO and ridge regression	Shape/firstorder GLCM/gray-level size zone matrix/gray-level distance zone matrix/neighborhood gray-tone difference matrix/gray-level run length matrix	Elastic net logistic regression	Diagnosis	AUC: 0.816
Li et al. (33)	Prospective Single center	Liver	144	BMUS/CEMF	472	Spearman’s correlation coefficient	Conventional radiomics/ORF/CEMF features	—	Diagnosis	Mean AUC: 0.78–0.85 (the multiparametric ultrasomics model)
Wang et al. (34)	Prospective Multicentre	Liver	654	SWE	—	CNN	—	Student’s t test/Mann–Whitney U test	Prognosis	AUC: 0.97 (F4) AUC: 0.98 (F3) AUC: 0.85 (F2)
Hu et al. (38)	Retrospective Multicentre	Liver	482	CEUS	1044	LASSO	—	LASSO	Prognosis	AUC: 0.731 p = 0.015
Liang et al. (39)	Retrospective Multicentre	Thyroid	137	BMUS	1044	LASSO	—	Univariate logistic regression	Diagnosis	AUC: 0.921 (training cohort) AUC: 0.931 (validation cohort)
Liu et al. (40)	Retrospective Single center	Lymph node	1216	BMUS	614	combined feature selection strategy	Echo/posterior acoustic/calcification	—	Prognosis	AUC: 0.782
Park et al. (41)	Retrospective Single center	Thyroid	768	BMUS	730	LASSO	—	LASSO/Cox regression	Prognosis	C-index: 0.777; 95%[CI]: 0.735, 0.829
Liu et al. (42)	Retrospective Single center	Lymph node	75	BMUS/SE-US	684	SVM	—	Delong’s test	Prognosis	AUC: 0.90 Accuracy: 0.85 Sensitivity: 0.77 Specificity: 0.88

The design of the studies, category of tumors, number of patients, number of features, type of features, mode build method, endpoint, diagnostic modality, and results of the studies were considered. The name of the first author and the reference number are indicated in the first column. BMUS, B-mode ultrasound; SWE, shear wave elastography; SWV, shear wave viscosity; CEUS, contrast-enhanced ultrasound; SE-US, strain ultrasound elastography; GM, the gray-scale modality; GEM, the gray-scale and elastography modality; GEVM, gray-scale, elastography and viscosity modality; SVM, support vector machine; LASSO, least absolute shrinkage and selection operator; RLM, gray level run-length matrix; CEMF, contrast-enhanced micro-flow; ORF, original radiofrequency; CEMF, contrast-enhanced micro-flow; and SRT, sparse representation theory.