Tags

Type your tag names separated by a space and hit enter

Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children.

Abstract

Importance

Duodenal biopsies from children with enteropathies associated with undernutrition, such as environmental enteropathy (EE) and celiac disease (CD), display significant histopathological overlap.

Objective

To develop a convolutional neural network (CNN) to enhance the detection of pathologic morphological features in diseased vs healthy duodenal tissue.

Design, Setting, and Participants

In this prospective diagnostic study, a CNN consisting of 4 convolutions, 1 fully connected layer, and 1 softmax layer was trained on duodenal biopsy images. Data were provided by 3 sites: Aga Khan University Hospital, Karachi, Pakistan; University Teaching Hospital, Lusaka, Zambia; and University of Virginia, Charlottesville. Duodenal biopsy slides from 102 children (10 with EE from Aga Khan University Hospital, 16 with EE from University Teaching Hospital, 34 with CD from University of Virginia, and 42 with no disease from University of Virginia) were converted into 3118 images. The CNN was designed and analyzed at the University of Virginia. The data were collected, prepared, and analyzed between November 2017 and February 2018.

Main Outcomes and Measures

Classification accuracy of the CNN per image and per case and incorrect classification rate identified by aggregated 10-fold cross-validation confusion/error matrices of CNN models.

Results

Overall, 102 children participated in this study, with a median (interquartile range) age of 31.0 (20.3-75.5) months and a roughly equal sex distribution, with 53 boys (51.9%). The model demonstrated 93.4% case-detection accuracy and had a false-negative rate of 2.4%. Confusion metrics indicated most incorrect classifications were between patients with CD and healthy patients. Feature map activations were visualized and learned distinctive patterns, including microlevel features in duodenal tissues, such as alterations in secretory cell populations.

Conclusions and Relevance

A machine learning-based histopathological analysis model demonstrating 93.4% classification accuracy was developed for identifying and differentiating between duodenal biopsies from children with EE and CD. The combination of the CNN with a deconvolutional network enabled feature recognition and highlighted secretory cells' role in the model's ability to differentiate between these histologically similar diseases.

Links

  • PMC Free PDF
  • PMC Free Full Text
  • Publisher Full Text
  • Authors+Show Affiliations

    ,

    Division of Pediatric Gastroenterology and Hepatology, Department of Pediatrics, University of Virginia, Charlottesville. Department of Pediatrics and Child Health, Aga Khan University, Karachi, Pakistan.

    ,

    Systems and Information Engineering, University of Virginia, Charlottesville.

    ,

    Division of Pediatric Gastroenterology and Hepatology, Department of Pediatrics, University of Virginia, Charlottesville.

    ,

    Department of Pediatrics and Child Health, Aga Khan University, Karachi, Pakistan.

    ,

    Department of Pediatrics and Child Health, Aga Khan University, Karachi, Pakistan.

    ,

    Deparment of Pathology, University of Virginia, Charlottesville.

    ,

    Blizard Institute, Barts and The London School of Medicine, Queen Mary University of London, London, United Kingdom. Tropical Gastroenterology and Nutrition Group, University of Zambia School of Medicine, Lusaka, Zambia.

    ,

    Tropical Gastroenterology and Nutrition Group, University of Zambia School of Medicine, Lusaka, Zambia.

    ,

    Department of Pediatrics and Child Health, Aga Khan University, Karachi, Pakistan.

    ,

    Division of Pediatric Gastroenterology and Hepatology, Department of Pediatrics, University of Virginia, Charlottesville.

    Data Science Institute, University of Virginia, Charlottesville.

    Source

    JAMA network open 2:6 2019 Jun 05 pg e195822

    Pub Type(s)

    Journal Article

    Language

    eng

    PubMed ID

    31199451

    Citation

    Syed, Sana, et al. "Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children." JAMA Network Open, vol. 2, no. 6, 2019, pp. e195822.
    Syed S, Al-Boni M, Khan MN, et al. Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children. JAMA Netw Open. 2019;2(6):e195822.
    Syed, S., Al-Boni, M., Khan, M. N., Sadiq, K., Iqbal, N. T., Moskaluk, C. A., ... Brown, D. E. (2019). Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children. JAMA Network Open, 2(6), pp. e195822. doi:10.1001/jamanetworkopen.2019.5822.
    Syed S, et al. Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children. JAMA Netw Open. 2019 Jun 5;2(6):e195822. PubMed PMID: 31199451.
    * Article titles in AMA citation format should be in sentence-case
    TY - JOUR T1 - Assessment of Machine Learning Detection of Environmental Enteropathy and Celiac Disease in Children. AU - Syed,Sana, AU - Al-Boni,Mohammad, AU - Khan,Marium N, AU - Sadiq,Kamran, AU - Iqbal,Najeeha T, AU - Moskaluk,Christopher A, AU - Kelly,Paul, AU - Amadi,Beatrice, AU - Ali,S Asad, AU - Moore,Sean R, AU - Brown,Donald E, Y1 - 2019/06/05/ PY - 2019/6/15/entrez PY - 2019/6/15/pubmed PY - 2019/6/15/medline SP - e195822 EP - e195822 JF - JAMA network open JO - JAMA Netw Open VL - 2 IS - 6 N2 - Importance: Duodenal biopsies from children with enteropathies associated with undernutrition, such as environmental enteropathy (EE) and celiac disease (CD), display significant histopathological overlap. Objective: To develop a convolutional neural network (CNN) to enhance the detection of pathologic morphological features in diseased vs healthy duodenal tissue. Design, Setting, and Participants: In this prospective diagnostic study, a CNN consisting of 4 convolutions, 1 fully connected layer, and 1 softmax layer was trained on duodenal biopsy images. Data were provided by 3 sites: Aga Khan University Hospital, Karachi, Pakistan; University Teaching Hospital, Lusaka, Zambia; and University of Virginia, Charlottesville. Duodenal biopsy slides from 102 children (10 with EE from Aga Khan University Hospital, 16 with EE from University Teaching Hospital, 34 with CD from University of Virginia, and 42 with no disease from University of Virginia) were converted into 3118 images. The CNN was designed and analyzed at the University of Virginia. The data were collected, prepared, and analyzed between November 2017 and February 2018. Main Outcomes and Measures: Classification accuracy of the CNN per image and per case and incorrect classification rate identified by aggregated 10-fold cross-validation confusion/error matrices of CNN models. Results: Overall, 102 children participated in this study, with a median (interquartile range) age of 31.0 (20.3-75.5) months and a roughly equal sex distribution, with 53 boys (51.9%). The model demonstrated 93.4% case-detection accuracy and had a false-negative rate of 2.4%. Confusion metrics indicated most incorrect classifications were between patients with CD and healthy patients. Feature map activations were visualized and learned distinctive patterns, including microlevel features in duodenal tissues, such as alterations in secretory cell populations. Conclusions and Relevance: A machine learning-based histopathological analysis model demonstrating 93.4% classification accuracy was developed for identifying and differentiating between duodenal biopsies from children with EE and CD. The combination of the CNN with a deconvolutional network enabled feature recognition and highlighted secretory cells' role in the model's ability to differentiate between these histologically similar diseases. SN - 2574-3805 UR - https://www.unboundmedicine.com/medline/citation/31199451/Assessment_of_Machine_Learning_Detection_of_Environmental_Enteropathy_and_Celiac_Disease_in_Children L2 - https://jamanetwork.com/journals/jamanetworkopen/fullarticle/10.1001/jamanetworkopen.2019.5822 DB - PRIME DP - Unbound Medicine ER -