A new xAI framework with feature explainability for tumors decision-making in Ultrasound data: comparing with Grad-CAM

Song, Di1,2; Yao, Jincao3; Jiang, Yitao4; Shi, Siyuan5; Cui, Chen5; Wang, Liping3; Wang, Lijing3; Wu, Huaiyu1,2; Tian, Hongtian1,2; Ye, Xiuqin1,2; Ou, Di3; Li, Wei3; Feng, Na3; Pan, Weiyun3; Song, Mei3; Xu, Jinfeng1,2; Xu, Dong3; Wu, Linghu1,2; Dong, Fajin1,2

2023-06-01

10.1016/j.cmpb.2023.107527

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE   Impact Factor & Quartile

0169-2607

1872-7565

Background and objective: The value of implementing artificial intelligence (AI) on ultrasound screening for thyroid cancer has been acknowledged, with numerous early studies confirming AI might help physi-cians acquire more accurate diagnoses. However, the black box nature of AI's decision-making process makes it difficult for users to grasp the foundation of AI's predictions. Furthermore, explainability is not only related to AI performance, but also responsibility and risk in medical diagnosis. In this paper, we offer Explainer, an intrinsically explainable framework that can categorize images and create heatmaps highlighting the regions on which its prediction is based. Methods: A dataset of 19341 thyroid ultrasound images with pathological results and physician -annotated TI-RADS features is used to train and test the robustness of the proposed framework. Then we conducted a benign-malignant classification study to determine whether physicians perform better with the assistance of an explainer than they do alone or with Gradient-weighted Class Activation Mapping (Grad-CAM). Results: Reader studies show that the Explainer can achieve a more accurate diagnosis while explaining heatmaps, and that physicians' performances are improved when assisted by the Explainer. Case study results confirm that the Explainer is capable of locating more reasonable and feature-related regions than the Grad-CAM. Conclusions: The Explainer offers physicians a tool to understand the basis of AI predictions and evaluate their reliability, which has the potential to unbox the "black box" of medical imaging AI. (c) 2023 Elsevier B.V. All rights reserved.

Explainability Artificial intelligence Medical decision support Ultrasound Thyroid cancer

SCI

English

Corresponding

Commission of Science and Tech- nology of Shenzhen[GJHZ20200731095401004]

Computer Science ; Engineering ; Medical Informatics

Computer Science, Interdisciplinary Applications ; Computer Science, Theory & Methods ; Engineering, Biomedical ; Medical Informatics

WOS:000983668100001

ELSEVIER IRELAND LTD

COMPUTER SCIENCE

Web of Science

1.Jinan Univ, Shenzhen Peoples Hosp, Clin Med Coll 2, Dept Ultrasound, Shenzhen 518020, Guangdong, Peoples R China
2.Southern Univ Sci & Technol, Affiliated Hosp 1, Shenzhen 518020, Guangdong, Peoples R China
3.Univ Chinese Acad Sci, Zhejiang Canc Hosp, Inst Basic Med & Canc IBMC, Chinese Acad Sci,Canc Hosp, Hangzhou 310022, Zhejiang, Peoples R China
4.Microport Prophecy, Res & Dev Dept, Shanghai 201203, Peoples R China
5.Illuminate LLC, Res & Dev Dept, Shenzhen 518000, Guangdong, Peoples R China

Song, Di,Yao, Jincao,Jiang, Yitao,et al. A new xAI framework with feature explainability for tumors decision-making in Ultrasound data: comparing with Grad-CAM[J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE,2023,235.

Song, Di.,Yao, Jincao.,Jiang, Yitao.,Shi, Siyuan.,Cui, Chen.,...&Dong, Fajin.(2023).A new xAI framework with feature explainability for tumors decision-making in Ultrasound data: comparing with Grad-CAM.COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE,235.

Song, Di,et al."A new xAI framework with feature explainability for tumors decision-making in Ultrasound data: comparing with Grad-CAM".COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE 235(2023).