Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data
Purpose To develop a model for predicting response to total neoadjuvant treatment (TNT) for patients with locally advanced rectal cancer (LARC) based on baseline magnetic resonance imaging (MRI) and clinical data using artificial intelligence methods. Methods Baseline MRI and clinical data were cura...
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SAGE Publishing
2023-07-01
|
| Series: | Technology in Cancer Research & Treatment |
| Online Access: | https://doi.org/10.1177/15330338231186467 |
| _version_ | 1797783251643793408 |
|---|---|
| author | Ganlu Ouyang MD Zhebin Chen PhD Meng Dou PhD Xu Luo PhD Han Wen PhD Xiangbing Deng PhD Wenjian Meng PhD Yongyang Yu PhD Bing Wu PhD Dan Jiang PhD Ziqiang Wang PhD Yu Yao PhD Xin Wang PhD |
| author_facet | Ganlu Ouyang MD Zhebin Chen PhD Meng Dou PhD Xu Luo PhD Han Wen PhD Xiangbing Deng PhD Wenjian Meng PhD Yongyang Yu PhD Bing Wu PhD Dan Jiang PhD Ziqiang Wang PhD Yu Yao PhD Xin Wang PhD |
| author_sort | Ganlu Ouyang MD |
| collection | DOAJ |
| description | Purpose To develop a model for predicting response to total neoadjuvant treatment (TNT) for patients with locally advanced rectal cancer (LARC) based on baseline magnetic resonance imaging (MRI) and clinical data using artificial intelligence methods. Methods Baseline MRI and clinical data were curated from patients with LARC and analyzed using logistic regression (LR) and deep learning (DL) methods to predict TNT response retrospectively. We defined two groups of response to TNT as pathological complete response (pCR) versus non-pCR (Group 1), and high sensitivity [tumor regression grade (TRG) 0 and TRG 1] versus moderate sensitivity (TRG 2 or patients with TRG 3 and a reduction in tumor volume of at least 20% compared to baseline) versus low sensitivity (TRG 3 and a reduction in tumor volume <20% compared to baseline) (Group 2). We extracted and selected clinical and radiomic features on baseline T2WI. Then we built LR models and DL models. Receiver operating characteristic (ROC) curves analysis was performed to assess predictive performance of models. Results Eighty-nine patients were assigned to the training cohort, and 29 patients were assigned to the testing cohort. The area under receiver operating characteristics curve (AUC) of LR models, which were predictive of high sensitivity and pCR, were 0.853 and 0.866, respectively. Whereas the AUCs of DL models were 0.829 and 0.838, respectively. After 10 rounds of cross validation, the accuracy of the models in Group 1 is higher than in Group 2. Conclusion There was no significant difference between LR model and DL model. Artificial Intelligence-based radiomics biomarkers may have potential clinical implications for adaptive and personalized therapy. |
| first_indexed | 2024-03-13T00:23:24Z |
| format | Article |
| id | doaj.art-99a3c157b812464f8ce95e33b5baeb93 |
| institution | Directory Open Access Journal |
| issn | 1533-0338 |
| language | English |
| last_indexed | 2024-03-13T00:23:24Z |
| publishDate | 2023-07-01 |
| publisher | SAGE Publishing |
| record_format | Article |
| series | Technology in Cancer Research & Treatment |
| spelling | doaj.art-99a3c157b812464f8ce95e33b5baeb932023-07-11T09:33:35ZengSAGE PublishingTechnology in Cancer Research & Treatment1533-03382023-07-012210.1177/15330338231186467Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical DataGanlu Ouyang MD0Zhebin Chen PhD1Meng Dou PhD2Xu Luo PhD3Han Wen PhD4Xiangbing Deng PhD5Wenjian Meng PhD6Yongyang Yu PhD7Bing Wu PhD8Dan Jiang PhD9Ziqiang Wang PhD10Yu Yao PhD11Xin Wang PhD12 Lung Cancer Center, , Chengdu, China University of Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China Department of Gastrointestinal Surgery, , Chengdu, China Department of Gastrointestinal Surgery, , Chengdu, China Department of Gastrointestinal Surgery, , Chengdu, China Department of Radiology, , Chengdu, China Department of Pathology, , Chengdu, China Department of Gastrointestinal Surgery, , Chengdu, China University of Chinese Academy of Sciences, Beijing, China Department of Abdominal Oncology, Cancer Center, , Chengdu, ChinaPurpose To develop a model for predicting response to total neoadjuvant treatment (TNT) for patients with locally advanced rectal cancer (LARC) based on baseline magnetic resonance imaging (MRI) and clinical data using artificial intelligence methods. Methods Baseline MRI and clinical data were curated from patients with LARC and analyzed using logistic regression (LR) and deep learning (DL) methods to predict TNT response retrospectively. We defined two groups of response to TNT as pathological complete response (pCR) versus non-pCR (Group 1), and high sensitivity [tumor regression grade (TRG) 0 and TRG 1] versus moderate sensitivity (TRG 2 or patients with TRG 3 and a reduction in tumor volume of at least 20% compared to baseline) versus low sensitivity (TRG 3 and a reduction in tumor volume <20% compared to baseline) (Group 2). We extracted and selected clinical and radiomic features on baseline T2WI. Then we built LR models and DL models. Receiver operating characteristic (ROC) curves analysis was performed to assess predictive performance of models. Results Eighty-nine patients were assigned to the training cohort, and 29 patients were assigned to the testing cohort. The area under receiver operating characteristics curve (AUC) of LR models, which were predictive of high sensitivity and pCR, were 0.853 and 0.866, respectively. Whereas the AUCs of DL models were 0.829 and 0.838, respectively. After 10 rounds of cross validation, the accuracy of the models in Group 1 is higher than in Group 2. Conclusion There was no significant difference between LR model and DL model. Artificial Intelligence-based radiomics biomarkers may have potential clinical implications for adaptive and personalized therapy.https://doi.org/10.1177/15330338231186467 |
| spellingShingle | Ganlu Ouyang MD Zhebin Chen PhD Meng Dou PhD Xu Luo PhD Han Wen PhD Xiangbing Deng PhD Wenjian Meng PhD Yongyang Yu PhD Bing Wu PhD Dan Jiang PhD Ziqiang Wang PhD Yu Yao PhD Xin Wang PhD Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data Technology in Cancer Research & Treatment |
| title | Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data |
| title_full | Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data |
| title_fullStr | Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data |
| title_full_unstemmed | Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data |
| title_short | Predicting Rectal Cancer Response to Total Neoadjuvant Treatment Using an Artificial Intelligence Model Based on Magnetic Resonance Imaging and Clinical Data |
| title_sort | predicting rectal cancer response to total neoadjuvant treatment using an artificial intelligence model based on magnetic resonance imaging and clinical data |
| url | https://doi.org/10.1177/15330338231186467 |
| work_keys_str_mv | AT ganluouyangmd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT zhebinchenphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT mengdouphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT xuluophd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT hanwenphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT xiangbingdengphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT wenjianmengphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT yongyangyuphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT bingwuphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT danjiangphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT ziqiangwangphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT yuyaophd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata AT xinwangphd predictingrectalcancerresponsetototalneoadjuvanttreatmentusinganartificialintelligencemodelbasedonmagneticresonanceimagingandclinicaldata |