Interpretable Artificial Intelligence in Assisting Treatment Response Prediction for Locally Advanced Rectal Cancer After Neoadjuvant Chemoradiotherapy: A Prospective, Multicenter, Human-Model Interaction Study.
Authors
Affiliations (11)
Affiliations (11)
- Department of Radiation Oncology, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Biomedical Innovation Center, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Colorectal and Pelvic Floor Diseases, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; State Key Laboratory of Metabolic Dysregulation & Prevention and Treatment of Esophageal Cancer, Tianjian Laboratory of Advanced Biomedical Sciences, Academy of Medical Sciences, Zhengzhou University, Zhengzhou, Henan, China.
- Department of Radiology, Guangdong Provincial People's Hospital, Southern Medical University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, Guangdong, China.
- Department of Radiology, Jiaxing Traditional Chinese Medicine Hospital Affiliated to Zhejiang Chinese Medical University, Jiaxing, Zhejiang, China.
- Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong, China; State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong, China; HeNan Provincial Key Laboratory of Radiation Medicine, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China.
- Biomedical Innovation Center, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Colorectal and Pelvic Floor Diseases, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Department of Radiology, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China.
- Department of Radiation Oncology, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Biomedical Innovation Center, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Colorectal and Pelvic Floor Diseases, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China.
- Department of Medical Oncology, Sir Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, Zhejiang, China.
- The First School of Clinical Medicine, Zhejiang Chinese Medical University, Hangzhou, Zhejiang, China.
- Department of Radiology, Guangdong Provincial People's Hospital, Southern Medical University, Guangzhou, Guangdong, China; Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangzhou, Guangdong, China. Electronic address: [email protected].
- State Key Laboratory of Metabolic Dysregulation & Prevention and Treatment of Esophageal Cancer, Tianjian Laboratory of Advanced Biomedical Sciences, Academy of Medical Sciences, Zhengzhou University, Zhengzhou, Henan, China; HeNan Provincial Key Laboratory of Radiation Medicine, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China; Department of Pathology, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China. Electronic address: [email protected].
- Department of Radiation Oncology, the Sixth Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong, China; State Key Laboratory of Metabolic Dysregulation & Prevention and Treatment of Esophageal Cancer, Tianjian Laboratory of Advanced Biomedical Sciences, Academy of Medical Sciences, Zhengzhou University, Zhengzhou, Henan, China; HeNan Provincial Key Laboratory of Radiation Medicine, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China; Department of Radiation Oncology, the First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan, China. Electronic address: [email protected].
Abstract
Preoperative assessment of pathologic complete response (pCR) to neoadjuvant therapy is an urgent need for anorectal preservation in patients with locally advanced rectal cancer (LARC). Artificial intelligence assistance remains challenging due to a lack of prospective validation and reliable interpretability. Eligible patients with LARC were retrospectively collected. Radiomic features extracted from postneoadjuvant therapy magnetic resonance imaging were applied to train a Deep Residual Shrinkage Network (DRSN) to generate Radscore for pCR probability. DRSN was integrated with significant clinicopathological factors to construct a multimodality model, named as RAPIDS-II, in the training set. RAPIDS-II performance in pCR prediction was verified in a testing set and further confirmed in a multicenter, prospective validation trial (NCT number: 04278274). The improvements of radiologists' visual assessment with RAPIDS-II assistance were evaluated in this prospective cohort. Area under curve (AUC) was used as primary endpoint for model performance. Retrospectively recruited 823 patients with LARC were divided into the training set (n = 575) and the testing set (n = 248). Compared with the DRSN model, RAPIDS-II showed a comparable AUC of 0.813 (95% CI, 0.736-0.874) in the testing set (P = 0.020). In the prospective validation cohort (n = 207), RAPIDS-II performed robustly with AUC of 0.795 (95%CI, 0.723-0.859) in identifying patients with pCR. Importantly, RAPIDS-II assistance improved in overall AUC and sensitivity of radiologists' visual assessment, especially for junior radiologists. Interpretable SHapley Additive exPlanations analysis identified that Radscore attributed most to RAPIDS-II prediction. The interpretable RAPIDS-II model demonstrates good performance in pCR evaluation and shows potential as a tool to assist clinicians, particularly those with less experience, in tailoring individualized therapy.