Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy
Device-assisted enteroscopy (DAE) is capable of evaluating the entire gastrointestinal tract, identifying multiple lesions. Nevertheless, DAE’s diagnostic yield is suboptimal. Convolutional neural networks (CNN) are multi-layer architecture artificial intelligence models suitable for image analysis,...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Cancers |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-6694/16/1/208 |
| _version_ | 1797359032262983680 |
|---|---|
| author | Francisco Mendes Miguel Mascarenhas Tiago Ribeiro João Afonso Pedro Cardoso Miguel Martins Hélder Cardoso Patrícia Andrade João P. S. Ferreira Miguel Mascarenhas Saraiva Guilherme Macedo |
| author_facet | Francisco Mendes Miguel Mascarenhas Tiago Ribeiro João Afonso Pedro Cardoso Miguel Martins Hélder Cardoso Patrícia Andrade João P. S. Ferreira Miguel Mascarenhas Saraiva Guilherme Macedo |
| author_sort | Francisco Mendes |
| collection | DOAJ |
| description | Device-assisted enteroscopy (DAE) is capable of evaluating the entire gastrointestinal tract, identifying multiple lesions. Nevertheless, DAE’s diagnostic yield is suboptimal. Convolutional neural networks (CNN) are multi-layer architecture artificial intelligence models suitable for image analysis, but there is a lack of studies about their application in DAE. Our group aimed to develop a multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. In total, 338 exams performed in two specialized centers were retrospectively evaluated, with 152 single-balloon enteroscopies (Fujifilm®, Porto, Portugal), 172 double-balloon enteroscopies (Olympus<sup>®</sup>, Porto, Portugal) and 14 motorized spiral enteroscopies (Olympus<sup>®</sup>, Porto, Portugal); then, 40,655 images were divided in a training dataset (90% of the images, <i>n</i> = 36,599) and testing dataset (10% of the images, <i>n</i> = 4066) used to evaluate the model. The CNN’s output was compared to an expert consensus classification. The model was evaluated by its sensitivity, specificity, positive (PPV) and negative predictive values (NPV), accuracy and area under the precision recall curve (AUC-PR). The CNN had an 88.9% sensitivity, 98.9% specificity, 95.8% PPV, 97.1% NPV, 96.8% accuracy and an AUC-PR of 0.97. Our group developed the first multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. The development of accurate deep learning models is of utmost importance for increasing the diagnostic yield of DAE-based panendoscopy. |
| first_indexed | 2024-03-08T15:10:48Z |
| format | Article |
| id | doaj.art-d9778536867c4723bd1312fae95c78f9 |
| institution | Directory Open Access Journal |
| issn | 2072-6694 |
| language | English |
| last_indexed | 2024-03-08T15:10:48Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Cancers |
| spelling | doaj.art-d9778536867c4723bd1312fae95c78f92024-01-10T14:53:02ZengMDPI AGCancers2072-66942024-01-0116120810.3390/cancers16010208Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted EnteroscopyFrancisco Mendes0Miguel Mascarenhas1Tiago Ribeiro2João Afonso3Pedro Cardoso4Miguel Martins5Hélder Cardoso6Patrícia Andrade7João P. S. Ferreira8Miguel Mascarenhas Saraiva9Guilherme Macedo10Alameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalDepartment of Mechanical Engineering, Faculty of Engineering, University of Porto, Rua Dr. Roberto Frias, 4200-465 Porto, PortugalManopH Gastroenterology Clinic, R. de Sá da Bandeira 752, 4000-432 Porto, PortugalAlameda Professor Hernâni Monteiro, Department of Gastroenterology, São João University Hospital, 4200-427 Porto, PortugalDevice-assisted enteroscopy (DAE) is capable of evaluating the entire gastrointestinal tract, identifying multiple lesions. Nevertheless, DAE’s diagnostic yield is suboptimal. Convolutional neural networks (CNN) are multi-layer architecture artificial intelligence models suitable for image analysis, but there is a lack of studies about their application in DAE. Our group aimed to develop a multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. In total, 338 exams performed in two specialized centers were retrospectively evaluated, with 152 single-balloon enteroscopies (Fujifilm®, Porto, Portugal), 172 double-balloon enteroscopies (Olympus<sup>®</sup>, Porto, Portugal) and 14 motorized spiral enteroscopies (Olympus<sup>®</sup>, Porto, Portugal); then, 40,655 images were divided in a training dataset (90% of the images, <i>n</i> = 36,599) and testing dataset (10% of the images, <i>n</i> = 4066) used to evaluate the model. The CNN’s output was compared to an expert consensus classification. The model was evaluated by its sensitivity, specificity, positive (PPV) and negative predictive values (NPV), accuracy and area under the precision recall curve (AUC-PR). The CNN had an 88.9% sensitivity, 98.9% specificity, 95.8% PPV, 97.1% NPV, 96.8% accuracy and an AUC-PR of 0.97. Our group developed the first multidevice CNN for panendoscopic detection of clinically relevant lesions during DAE. The development of accurate deep learning models is of utmost importance for increasing the diagnostic yield of DAE-based panendoscopy.https://www.mdpi.com/2072-6694/16/1/208artificial intelligencedeep learningpanendoscopydevice-assisted enteroscopy |
| spellingShingle | Francisco Mendes Miguel Mascarenhas Tiago Ribeiro João Afonso Pedro Cardoso Miguel Martins Hélder Cardoso Patrícia Andrade João P. S. Ferreira Miguel Mascarenhas Saraiva Guilherme Macedo Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy Cancers artificial intelligence deep learning panendoscopy device-assisted enteroscopy |
| title | Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy |
| title_full | Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy |
| title_fullStr | Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy |
| title_full_unstemmed | Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy |
| title_short | Artificial Intelligence and Panendoscopy—Automatic Detection of Clinically Relevant Lesions in Multibrand Device-Assisted Enteroscopy |
| title_sort | artificial intelligence and panendoscopy automatic detection of clinically relevant lesions in multibrand device assisted enteroscopy |
| topic | artificial intelligence deep learning panendoscopy device-assisted enteroscopy |
| url | https://www.mdpi.com/2072-6694/16/1/208 |
| work_keys_str_mv | AT franciscomendes artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT miguelmascarenhas artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT tiagoribeiro artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT joaoafonso artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT pedrocardoso artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT miguelmartins artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT heldercardoso artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT patriciaandrade artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT joaopsferreira artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT miguelmascarenhassaraiva artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy AT guilhermemacedo artificialintelligenceandpanendoscopyautomaticdetectionofclinicallyrelevantlesionsinmultibranddeviceassistedenteroscopy |