Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 54 min ago
Peer-reviewedCC BY 4.0JMIR Medical Informatics

Automated Lewis Score Estimation for Small Bowel Crohn Disease Using Capsule Endoscopy: Retrospective Cohort Study

  • Yingying Meng1
  • Min Gao2
  • Jie Wu1
  • Ying Huang1
  1. 1Children's Hospital of Fudan University Shanghai China
  2. 2Chongqing Jinshan Technology (Group) Co, Ltd Chongqing China
Read full textDOICite

Abstract

Background: Small bowel capsule endoscopy (SBCE) is widely used to evaluate small bowel involvement in Crohn disease (CD); however, manual Lewis score calculation is time-consuming and susceptible to interobserver variability.

Objective: This study aimed to develop and preliminarily validate a multitask deep learning system for automated Lewis score estimation in patients with small bowel CD undergoing SBCE.

Methods: We developed Eval-Model, a multitask framework integrating You Only Look Once (YOLO)–style lesion detection and localization for ulcers and edema, a stenosis classification branch, and a long short-term memory–based temporal regression module. The primary dataset comprised SBCE examinations from 814 patients at the Children’s Hospital of Fudan University. In this dataset, 725 (89.1%) cases were included for lesion-level annotation, and 89 (10.9%) cases with expert-consensus Lewis scores were used for scoring model development and internal testing. These 89 cases comprised 49 (55.1%) training cases and 40 (44.9%) internal test cases. All splits were performed at the patient level. A retrospective same-center temporal validation cohort of 20 consecutive cases from 2026 and 5 control cases were used for preliminary assessments of temporal robustness and specificity.

Results: In the 40-case internal Lewis score test set, Eval-Model achieved a mean squared error of 14.693 (95% CI 8.72-22.15), root mean squared error of 3.833 (95% CI 2.95-4.71), mean absolute error of 2.9 (95% CI 2.1-3.8), R2 of 0.82 (95% CI 0.70-0.90), and explained variance of 0.83 (95% CI 0.72-0.91). For lesion recognition, the model achieved an overall mean average precision of 0.94 with an inference time of 15 ms per frame. In the 20-case same-center temporal validation cohort, the mean squared error was 26.848 (95% CI 12.51-45.33), the root mean squared error was 5.182 (95% CI 3.54-6.73), the mean absolute error was 3.3 (95% CI 2.2-4.6), R2 rounded to 1.00 (95% CI 0.98-1.00), and the explained variance rounded to 1.00 (95% CI 0.98-1.00). These temporal estimates require cautious interpretation because of the small, same-center cohort.

Conclusions: Eval-Model demonstrated preliminary feasibility for automated Lewis score estimation from SBCE in small bowel CD. The system may serve as a standardized adjunct to expert review, but larger multicenter prospective validation with independent external annotation is required before clinical deployment.

Related

  1. Artificial Intelligence Versus Conventional Scoring for Assessing Small Bowel Capsule Endoscopy Cleanliness in Crohn's Disease
    Raphaëlle Rouveyre, Catherine Le Berre, Romain Leenhardt +10
    United European Gastroenterology JournalPeer-reviewedCC BY-NC-ND 4.0Full textGeneral biologyGeneral medicineHAL
  2. A promptable foundation model enables automated multi-task dataset construction and real-time pose estimation in mice
    J. Cobos, S. Thirard, M. Belkaid +1
    bioRxivPreprintCC BYFull textZoology & animal behaviourbioRxiv
  3. Comparative analysis of CNN, Transformer, and Mamba-based models for automated eye disease classification from fundus images
    Md Ahanaf Arif Khan, Sangeeta Biswas
    PLOS OnePeer-reviewedCC BYFull textNeurosciencePhysiologyPLOSEurope PMC
  4. FDUNet-LRSA: efficient frequency-adaptive segmentation for automated crop disease inspection
    Ru Han, Lei Shu, Grzegorz Cielniak +6
    Frontiers in Plant SciencePeer-reviewedCC BY 4.0Full textPlant biologyCrossrefEurope PMCPMC
  5. Automated screening of Parkinson's Disease using Smooth Pursuit Eye Movements
    J. Camilo Puerta-Acevedo, Maria F. Alcala-Durand, Julián D. Arias-Londoño +3
    arXivPreprintCC BY 4.0Full textBioinformatics & computational biologyarXiv

Cite

BibTeXRIS
BibTeX
@article{meng2026automated,
  title = {{Automated Lewis Score Estimation for Small Bowel Crohn Disease Using Capsule Endoscopy: Retrospective Cohort Study}},
  author = {Yingying Meng and Min Gao and Jie Wu and Ying Huang},
  journal = {JMIR Medical Informatics},
  year = {2026},
  month = oct,
  doi = {10.2196/92416},
  url = {https://doi.org/10.2196/92416},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Automated Lewis Score Estimation for Small Bowel Crohn Disease Using Capsule Endoscopy: Retrospective Cohort Study
AU  - Yingying Meng
AU  - Min Gao
AU  - Jie Wu
AU  - Ying Huang
PY  - 2026
DA  - 2026/10/13
JO  - JMIR Medical Informatics
DO  - 10.2196/92416
UR  - https://doi.org/10.2196/92416
AB  - Background: Small bowel capsule endoscopy (SBCE) is widely used to evaluate small bowel involvement in Crohn disease (CD); however, manual Lewis score calculation is time-consuming and susceptible to interobserver variability.  Objective: This study aimed to develop and preliminarily validate a multitask deep learning system for automated Lewis score estimation in patients with small bowel CD undergoing SBCE.  Methods: We developed Eval-Model, a multitask framework integrating You Only Look Once (YOLO)–style lesion detection and localization for ulcers and edema, a stenosis classification branch, and a long short-term memory–based temporal regression module. The primary dataset comprised SBCE examinations from 814 patients at the Children’s Hospital of Fudan University. In this dataset, 725 (89.1%) cases were included for lesion-level annotation, and 89 (10.9%) cases with expert-consensus Lewis scores were used for scoring model development and internal testing. These 89 cases comprised 49 (55.1%) training cases and 40 (44.9%) internal test cases. All splits were performed at the patient level. A retrospective same-center temporal validation cohort of 20 consecutive cases from 2026 and 5 control cases were used for preliminary assessments of temporal robustness and specificity.  Results: In the 40-case internal Lewis score test set, Eval-Model achieved a mean squared error of 14.693 (95% CI 8.72-22.15), root mean squared error of 3.833 (95% CI 2.95-4.71), mean absolute error of 2.9 (95% CI 2.1-3.8), R2 of 0.82 (95% CI 0.70-0.90), and explained variance of 0.83 (95% CI 0.72-0.91). For lesion recognition, the model achieved an overall mean average precision of 0.94 with an inference time of 15 ms per frame. In the 20-case same-center temporal validation cohort, the mean squared error was 26.848 (95% CI 12.51-45.33), the root mean squared error was 5.182 (95% CI 3.54-6.73), the mean absolute error was 3.3 (95% CI 2.2-4.6), R2 rounded to 1.00 (95% CI 0.98-1.00), and the explained variance rounded to 1.00 (95% CI 0.98-1.00). These temporal estimates require cautious interpretation because of the small, same-center cohort.  Conclusions: Eval-Model demonstrated preliminary feasibility for automated Lewis score estimation from SBCE in small bowel CD. The system may serve as a standardized adjunct to expert review, but larger multicenter prospective validation with independent external annotation is required before clinical deployment.
SN  - 2291-9694
ER  -