Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


Thesis_Ishan_Manandhar_final.pdf (1.29 MB)

ETD Abstract Container

Abstract Header

Artificial intelligence-based clinical classification of diseases: Utilizing gut microbiota as a feature for supervised learning and diagnostic screening of inflammatory bowel diseases

Manandhar, Ishan
http://rave.ohiolink.edu/etdc/view?acc_num=mco1628163254153436

Abstract Details

2021, Master of Science in Biomedical Sciences (MSBS), University of Toledo, Biomedical Sciences (Bioinformatics and Proteomics/Genomics).
Advances in machine learning (ML)-based techniques has popularized wide applications of artificial intelligence (AI) in various fields ranging from robotics to medicine. In recent years, there has been a surge in the application of AI to research in cardiovascular medicine, which is largely driven by the availability of large-scale clinical and multi-omics datasets. Such applications are providing a new perspective for a better understanding of cardiovascular disease, which could be used to develop novel diagnostic and therapeutic strategies. For example, studies have shown that ML has a substantial potential for early diagnosis of different types of cardiovascular diseases, prediction of adverse disease outcomes such as heart failure, and development of newer and personalized treatments. In chapter 1, I present a review of the current literature on a wide range of AI applications, including machine learning, reinforcement learning and deep learning, in cardiovascular medicine. The contents of this chapter were peer-reviewed and accepted as a review article in Comprehensive Physiology. My contribution to this publication was to collect literature, read, compose various sections and compile the review for a comprehensive presentation. I share first-co-authorship with Sachin Aryal on this publication. While we were undertaking the work described in Chapter 1, the Physiology branch of the Joe laboratory had discovered gut microbiota as new entrants into the field of cardiovascular medicine. Specifically, the gut microbiome (which represents the genomes of gut microbiota) is now recognized as an important feature in hypertension, which is the single largest risk factor for all cardiovascular illnesses. Risk prediction for cardiovascular disease, as described in Chapter 1, is possible with machine-learning, but such machine-learning did not use microbiome as a feature. Therefore, it was intriguing for us to test the hypothesis that the gut microbiome represents a new feature that could differentiate subjects with and without cardiovascular disease. To test this hypothesis, we extracted gut microbiome data from the American Gut Project (AGP; http://americangut.org) and developed machine-learning algorithms to classify cardiovascular disease and non-cardiovascular disease groups. This work was published with my contributions as a co-author. Because contributions of the gut microbiome are not limited to cardiovascular diseases but encompasses other gut pathologies such as inflammatory bowel diseases (IBD), I focused my thesis work on applying supervised machine-learning using gut microbiota to classify humans with and without IBD and to differentiate humans with Crohn’s Disease and Ulcerative Colitis. Our overall goal was to address the current gaps in the clinics to diagnose IBD. Despite the availability of various diagnostic tests for IBD, misdiagnosis of IBD occurs frequently, and thus there is a clinical need to further improve the diagnosis of IBD. In chapter 2, we hypothesized that supervised ML could be used to analyze gut microbiome data for predictive diagnostics of IBD. To test our hypothesis, fecal 16S metagenomic data of 729 IBD and 700 non-IBD subjects from the American Gut Project were analyzed using five different ML algorithms. Fifty differential bacterial taxa were identified (LEfSe: LDA > 3) between the IBD and non-IBD groups, and ML classifications trained with these taxonomic features using random forest (RF) achieved a testing AUC of ~0.80. Next, we tested if operational taxonomic units (OTUs), instead of bacterial taxa, could be used as ML features for diagnostic classification of IBD. Top 500 high-variance OTUs were used for ML training and an improved testing AUC of ~0.82 (RF) was achieved. Lastly, we tested if supervised ML could be used for differentiating Crohn’s disease and ulcerative colitis. Using 331 Crohn’s disease and 141 ulcerative colitis samples, 117 differential bacterial taxa (LEfSe: LDA > 3) were identified, and the RF model trained with differential taxonomic features or high-variance OTU features achieved a testing AUC > 0.90. In summary, chapter 2 demonstrates the promising potential of artificial intelligence via supervised ML modeling for predictive diagnostics of IBD using gut microbiome data. This work is my first-authored publication, which has been peer-reviewed and published by the American Physiological Society Journal- GI & Liver Physiology.
Bina Joe, Dr. (Advisor)
Robert Blumenthal, Dr. (Committee Member)
Xi Cheng, Dr. (Committee Member)
75 p.
Bioinformatics; Biomedical Research
artificial intelligence, machine learning, deep learning, reinforcement learning, physiology, cardiovascular, inflammatory bowel disease, Crohn’s disease, ulcerative colitis, gut microbiome, diagnosis

Recommended Citations

Citations

  • Manandhar, I. (2021). Artificial intelligence-based clinical classification of diseases: Utilizing gut microbiota as a feature for supervised learning and diagnostic screening of inflammatory bowel diseases [Master's thesis, University of Toledo]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=mco1628163254153436

    APA Style (7th edition)

  • Manandhar, Ishan. Artificial intelligence-based clinical classification of diseases: Utilizing gut microbiota as a feature for supervised learning and diagnostic screening of inflammatory bowel diseases. 2021. University of Toledo, Master's thesis. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=mco1628163254153436.

    MLA Style (8th edition)

  • Manandhar, Ishan. "Artificial intelligence-based clinical classification of diseases: Utilizing gut microbiota as a feature for supervised learning and diagnostic screening of inflammatory bowel diseases." Master's thesis, University of Toledo, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=mco1628163254153436

    Chicago Manual of Style (17th edition)

mco1628163254153436
131
© 2021, some rights reserved.Creative Commons License Artificial intelligence-based clinical classification of diseases: Utilizing gut microbiota as a feature for supervised learning and diagnostic screening of inflammatory bowel diseases by Ishan Manandhar is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. Based on a work at etd.ohiolink.edu.
This open access ETD is published by University of Toledo Health Science Campus and OhioLINK.