Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network
Introduction: Pancreatitis is a potentially lethal adverse event of endoscopic transpapillary placement of a self-expandable metal stent (SEMS) for malignant biliary obstruction (MBO). Deep learning-based image recognition has not been investigated in predicting pancreatitis in this setting. Method...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Georg Thieme Verlag KG
|
| Series: | Endoscopy International Open |
| Online Access: | http://www.thieme-connect.de/DOI/DOI?10.1055/a-2298-0147 |
| _version_ | 1797227154586468352 |
|---|---|
| author | Tsuyoshi Hamada Koichiro Yasaka Yousuke Nakai Rintaro Fukuda Ryunosuke Hakuta Kazunaga Ishigaki Sachiko Kanai Kensaku Noguchi Hiroki Oyama Tomotaka Saito Tatsuya Sato Tatsunori Suzuki Naminatsu Takahara Hiroyuki Isayama Osamu Abe Mitsuhiro Fujishiro |
| author_facet | Tsuyoshi Hamada Koichiro Yasaka Yousuke Nakai Rintaro Fukuda Ryunosuke Hakuta Kazunaga Ishigaki Sachiko Kanai Kensaku Noguchi Hiroki Oyama Tomotaka Saito Tatsuya Sato Tatsunori Suzuki Naminatsu Takahara Hiroyuki Isayama Osamu Abe Mitsuhiro Fujishiro |
| author_sort | Tsuyoshi Hamada |
| collection | DOAJ |
| description | Introduction: Pancreatitis is a potentially lethal adverse event of endoscopic transpapillary placement of a self-expandable metal stent (SEMS) for malignant biliary obstruction (MBO). Deep learning-based image recognition has not been investigated in predicting pancreatitis in this setting.
Methods: We included 70 patients who received endoscopic placement of a SEMS for nonresectable distal MBO. We constructed a convolutional neural network (CNN) model for pancreatitis prediction using a series of preprocedural computed tomography images covering the whole pancreas (≥ 120,960 augmented images in total). We examined the additional effects of the CNN-based probabilities on the following machine learning models based on clinical parameters: logistic regression, support vector machine with a linear or RBF kernel, random forest classifier, and gradient boosting classifier. Model performance was assessed based on the area under the curve (AUC) in the receiver operating characteristic analysis, positive predictive value (PPV), accuracy, and specificity.
Results: The CNN model was associated with moderate levels of performance metrics: AUC, 0.67; PPV, 0.45; accuracy, 0.66; and specificity, 0.63. When added to the machine learning models, the CNN-based probabilities increased the performance metrics. The logistic regression model with the CNN-based probabilities had an AUC of 0.74, PPV of 0.85, accuracy of 0.83, and specificity of 0.96, compared to 0.72, 0.78, 0.77, and 0.96, respectively, without the probabilities.
Conclusions: The CNN-based model may increase the predictability for pancreatitis following endoscopic placement of a biliary SEMS. Our findings support the potential of deep learning technology in improving prognostic models in pancreatobiliary therapeutic endoscopy. |
| first_indexed | 2024-04-24T14:36:18Z |
| format | Article |
| id | doaj.art-54f7e52ddac64e2aab0cac5849c2873d |
| institution | Directory Open Access Journal |
| issn | 2364-3722 2196-9736 |
| language | English |
| last_indexed | 2024-04-24T14:36:18Z |
| publisher | Georg Thieme Verlag KG |
| record_format | Article |
| series | Endoscopy International Open |
| spelling | doaj.art-54f7e52ddac64e2aab0cac5849c2873d2024-04-02T23:49:18ZengGeorg Thieme Verlag KGEndoscopy International Open2364-37222196-973610.1055/a-2298-0147Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural networkTsuyoshi Hamada0https://orcid.org/0000-0002-3937-2755Koichiro Yasaka1Yousuke Nakai2https://orcid.org/0000-0001-7411-1385Rintaro Fukuda3Ryunosuke Hakuta4Kazunaga Ishigaki5Sachiko Kanai6Kensaku Noguchi7Hiroki Oyama8Tomotaka Saito9Tatsuya Sato10Tatsunori Suzuki11Naminatsu Takahara12Hiroyuki Isayama13https://orcid.org/0000-0002-6206-9236Osamu Abe14Mitsuhiro Fujishiro15Gastroenterology, The University of TokyoRadiology, The University of Tokyo, Bunkyo-ku, JapanEndoscopy and Endoscopic Surgery, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, Juntendo University, Bunkyo-ku, JapanRadiology, The University of Tokyo, Bunkyo-ku, JapanGastroenterology, The University of Tokyo, Bunkyo-ku, JapanIntroduction: Pancreatitis is a potentially lethal adverse event of endoscopic transpapillary placement of a self-expandable metal stent (SEMS) for malignant biliary obstruction (MBO). Deep learning-based image recognition has not been investigated in predicting pancreatitis in this setting. Methods: We included 70 patients who received endoscopic placement of a SEMS for nonresectable distal MBO. We constructed a convolutional neural network (CNN) model for pancreatitis prediction using a series of preprocedural computed tomography images covering the whole pancreas (≥ 120,960 augmented images in total). We examined the additional effects of the CNN-based probabilities on the following machine learning models based on clinical parameters: logistic regression, support vector machine with a linear or RBF kernel, random forest classifier, and gradient boosting classifier. Model performance was assessed based on the area under the curve (AUC) in the receiver operating characteristic analysis, positive predictive value (PPV), accuracy, and specificity. Results: The CNN model was associated with moderate levels of performance metrics: AUC, 0.67; PPV, 0.45; accuracy, 0.66; and specificity, 0.63. When added to the machine learning models, the CNN-based probabilities increased the performance metrics. The logistic regression model with the CNN-based probabilities had an AUC of 0.74, PPV of 0.85, accuracy of 0.83, and specificity of 0.96, compared to 0.72, 0.78, 0.77, and 0.96, respectively, without the probabilities. Conclusions: The CNN-based model may increase the predictability for pancreatitis following endoscopic placement of a biliary SEMS. Our findings support the potential of deep learning technology in improving prognostic models in pancreatobiliary therapeutic endoscopy.http://www.thieme-connect.de/DOI/DOI?10.1055/a-2298-0147 |
| spellingShingle | Tsuyoshi Hamada Koichiro Yasaka Yousuke Nakai Rintaro Fukuda Ryunosuke Hakuta Kazunaga Ishigaki Sachiko Kanai Kensaku Noguchi Hiroki Oyama Tomotaka Saito Tatsuya Sato Tatsunori Suzuki Naminatsu Takahara Hiroyuki Isayama Osamu Abe Mitsuhiro Fujishiro Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network Endoscopy International Open |
| title | Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| title_full | Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| title_fullStr | Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| title_full_unstemmed | Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| title_short | Computed tomography-based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| title_sort | computed tomography based prediction of pancreatitis following biliary metal stent placement with the convolutional neural network |
| url | http://www.thieme-connect.de/DOI/DOI?10.1055/a-2298-0147 |
| work_keys_str_mv | AT tsuyoshihamada computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT koichiroyasaka computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT yousukenakai computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT rintarofukuda computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT ryunosukehakuta computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT kazunagaishigaki computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT sachikokanai computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT kensakunoguchi computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT hirokioyama computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT tomotakasaito computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT tatsuyasato computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT tatsunorisuzuki computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT naminatsutakahara computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT hiroyukiisayama computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT osamuabe computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork AT mitsuhirofujishiro computedtomographybasedpredictionofpancreatitisfollowingbiliarymetalstentplacementwiththeconvolutionalneuralnetwork |