An Interpretable Machine Learning Model to Classify Coronary Bifurcation Lesions

We present an interpretable machine learning model for medical diagnosis called sparse high-order interaction model with rejection option (SHIMR). A decision tree explains to a patient the diagnosis with a long rule (i.e., conjunction of many intervals), while SHIMR employs a weighted sum of short rules. Using proteomics data of 151 subjects in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, SHIMR is shown to be as accurate as other non-interpretable methods (Sensitivity, SN = 0.84 ± 0.1, Specificity, SP = 0.69 ± 0.15 and Area Under the Curve, AUC = 0.86 ± 0.09). For clinical usage, SHIMR has a function to abstain from making any diagnosis when it is not confident enough, so that a medical doctor can choose more accurate but invasive and/or more costly pathologies. The incorporation of a rejection option complements SHIMR in designing a multistage cost-effective diagnosis framework. Using a baseline concentration of cerebrospinal fluid (CSF) and plasma proteins from a common cohort of 141 subjects, SHIMR is shown to be effective in designing a patient-specific cost-effective Alzheimer’s disease (AD) pathology. Thus, interpretability, reliability and having the potential to design a patient-specific multistage cost-effective diagnosis framework can make SHIMR serve as an indispensable tool in the era of precision medicine that can cater to the demand of both doctors and patients, and reduce the overwhelming financial burden of medical diagnosis.

Download Full-text

AN INTERPRETABLE MACHINE LEARNING MODEL FOR INDIVIDUALIZED PROTOCOL SELECTION AND GONADOTROPIN DOSE SELECTION DURING OVARIAN STIMULATION

Fertility and Sterility ◽

10.1016/j.fertnstert.2021.07.479 ◽

2021 ◽

Vol 116 (3) ◽

pp. e174

Author(s):

Kevin E. Loewke ◽

Veronica I. Nutting ◽

Justina Hyunjii Cho ◽

David I. Hoffman ◽

Louis N. Weckstein ◽

...

Keyword(s):

Machine Learning ◽

Ovarian Stimulation ◽

Learning Model ◽

Dose Selection ◽

Interpretable Machine Learning ◽

Machine Learning Model ◽

Protocol Selection

Download Full-text

A Multi-Omics Interpretable Machine Learning Model Reveals Modes of Action of Small Molecules

Scientific Reports ◽

10.1038/s41598-020-57691-7 ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Natasha L. Patel-Murray ◽

Miriam Adam ◽

Nhan Huynh ◽

Brook T. Wassie ◽

Pamela Milani ◽

...

Keyword(s):

Machine Learning ◽

Small Molecules ◽

Learning Model ◽

Modes Of Action ◽

Interpretable Machine Learning ◽

Machine Learning Model

Download Full-text

An Interpretable Machine Learning Model for Human Fall Detection Systems Using Hybrid Intelligent Models

Challenges and Trends in Multimodal Fall Detection for Healthcare - Studies in Systems, Decision and Control ◽

10.1007/978-3-030-38748-8_8 ◽

2020 ◽

pp. 181-205 ◽

Cited By ~ 2

Author(s):

Paulo Vitor C. Souza ◽

Augusto J. Guimaraes ◽

Vanessa S. Araujo ◽

Lucas O. Batista ◽

Thiago S. Rezende

Keyword(s):

Machine Learning ◽

Fall Detection ◽

Learning Model ◽

Detection Systems ◽

Interpretable Machine Learning ◽

Machine Learning Model ◽

Intelligent Models

Download Full-text

Accelerated Reactivity Mechanism and Interpretable Machine Learning Model of N-Sulfonylimines toward Fast Multicomponent Reactions

Organic Letters ◽

10.1021/acs.orglett.0c03083 ◽

2020 ◽

Vol 22 (21) ◽

pp. 8480-8486

Author(s):

Krupal P. Jethava ◽

Jonathan Fine ◽

Yingqi Chen ◽

Ahad Hossain ◽

Gaurav Chopra

Keyword(s):

Machine Learning ◽

Multicomponent Reactions ◽

Learning Model ◽

Interpretable Machine Learning ◽

Machine Learning Model

Download Full-text

Interpretable Machine Learning Model for Mortality Prediction in ICU: A Multicenter Study

10.21203/rs.3.rs-83283/v1 ◽

2020 ◽

Author(s):

Ka Man Fong ◽

Shek Yin Au ◽

George Wing Yiu Ng ◽

Anne Kit Hung Leung

Keyword(s):

Machine Learning ◽

Hospital Mortality ◽

Severity Score ◽

Learning Model ◽

Mortality Prediction ◽

Large Dataset ◽

Interpretable Machine Learning ◽

Machine Learning Model ◽

Precision Recall Curve ◽

Apache Iv

Abstract Background: Researchers have long been struggling to improve the disease severity score in mortality prediction in ICU. The digitalization of medical health records and advancement of computation power have promoted the use of machine learning in critical care. This study aimed to develop an interpretable machine learning model using datasets from multicenters, and to compare with the APACHE IV, in predicting hospital mortality of patients admitted to ICU.Method: The datasets were assembled from the eICU database including 136145 patients across 208 hospitals throughout the U.S. and 5 ICUs in Hong Kong, including 10909 patients. The two datasets were first combined into one large dataset before 80:20 stratified split into the training set and the test set. The XGBoost machine algorithm was chosen to predict the hospital mortality. The variables in the model were the same as those included in the APACHE IV score. The discrimination and calibration of the model were assessed. The model would be interpreted using the Shapley Additive explanations values.Results: Of the 147054 patients in the whole cohort, the hospital mortality was 9.3%. The area under the precision-recall curve for the XGBoost algorithm was 0.57, and 0.49 for APACHE IV. Similarly, the XGBoost reached an area under the receiving operating curve (AUROC) of 0.90, while APACHE IV had an AUROC of 0.87. Additionally, the XGBoost algorithm showed better calibration than the APACHE IV. The three most important variables were age, heart rate, and whether the patient was on ventilator.Conclusions: The severity score developed by machine learning model using mutlicenter datasets outperformed the APACHE IV in predicting hospital mortality for patients admitted to ICU.

Download Full-text

Accelerated Reactivity Mechanism and Interpretable Machine Learning Model of N-Sulfonylimines Towards Fast Multicomponent Reactions

10.26434/chemrxiv.12116163.v1 ◽

2020 ◽

Cited By ~ 1

Author(s):

Krupal P. Jethava ◽

Jonathan A Fine ◽

Yingqi Chen ◽

Ahad Hossain ◽

Gaurav Chopra

Keyword(s):

Machine Learning ◽

Multicomponent Reaction ◽

Chemical Reactivity ◽

Reaction Times ◽

Learning Model ◽

General Strategy ◽

One Pot ◽

Interpretable Machine Learning ◽

Machine Learning Model ◽

Heterogeneous Reactivity

Predicting the outcome of chemical reactions using machine learning models has emerged as a promising research area in chemical science. However, the use of such models to prospectively test new reactions by interpreting chemical reactivity is limited. We have developed a new fast and one-pot multicomponent reaction of N-sulfonylimines with heterogenous reactivity. Fast reaction times (<5 min) for both acyclic and cyclic sulfonylimine encouraged us to investigate plausible reaction mechanisms using quantum mechanics to identify intermediates and transition states. The heterogeneous reactivity of N-sulfonylimine lead us to develop a human-interpretable machine learning model using positive and negative reaction profiles. We introduce chemical reactivity flowcharts to help chemists interpret the decisions made by the machine learning model for understanding heterogeneous reactivity of N-sulfonylimines. The model learns chemical patterns to accurately predict the reactivity of N-sulfonylimine with different carboxylic acids and can be used to suggest new reactions to elucidate the substrate scope of the reaction. We believe our human-interpretable machine learning approach is a general strategy that is useful to understand chemical reactivity of components for any multicomponent reaction to enhance synthesis of drug-like libraries.

Download Full-text

Accelerated Reactivity Mechanism and Interpretable Machine Learning Model of N-Sulfonylimines Towards Fast Multicomponent Reactions

10.26434/chemrxiv.12116163 ◽

2020 ◽

Author(s):

Krupal P. Jethava ◽

Jonathan A Fine ◽

Yingqi Chen ◽

Ahad Hossain ◽

Gaurav Chopra

Keyword(s):

Machine Learning ◽

Multicomponent Reaction ◽

Chemical Reactivity ◽

Reaction Times ◽

Learning Model ◽

General Strategy ◽

One Pot ◽

Interpretable Machine Learning ◽

Machine Learning Model ◽

Heterogeneous Reactivity

Predicting the outcome of chemical reactions using machine learning models has emerged as a promising research area in chemical science. However, the use of such models to prospectively test new reactions by interpreting chemical reactivity is limited. We have developed a new fast and one-pot multicomponent reaction of N-sulfonylimines with heterogenous reactivity. Fast reaction times (<5 min) for both acyclic and cyclic sulfonylimine encouraged us to investigate plausible reaction mechanisms using quantum mechanics to identify intermediates and transition states. The heterogeneous reactivity of N-sulfonylimine lead us to develop a human-interpretable machine learning model using positive and negative reaction profiles. We introduce chemical reactivity flowcharts to help chemists interpret the decisions made by the machine learning model for understanding heterogeneous reactivity of N-sulfonylimines. The model learns chemical patterns to accurately predict the reactivity of N-sulfonylimine with different carboxylic acids and can be used to suggest new reactions to elucidate the substrate scope of the reaction. We believe our human-interpretable machine learning approach is a general strategy that is useful to understand chemical reactivity of components for any multicomponent reaction to enhance synthesis of drug-like libraries.

Download Full-text