Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information
Aims: Metabolic dysfunction Associated Steatotic Liver Disease (MASLD) outcomes such as MASH (metabolic dysfunction associated steatohepatitis), fibrosis and cirrhosis are ordinarily determined by resource-intensive and invasive biopsies. We aim to show that routine clinical tests offer sufficient i...
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
Public Library of Science
2024
|
| _version_ | 1826313082138263552 |
|---|---|
| author | McTeer, M Applegate, D Mesenbrink, P Ratziu, V Schattenberg, JM Bugianesi, E Geier, A Romero Gomez, M Dufour, J Ekstedt, M Francque, S Yki-Jarvinen, H Allison, M Valenti, L Miele, L Pavlides, M Cobbold, J Papatheodoridis, G Holleboom, AG Tiniakos, D Brass, C Anstee, QM Missier, P |
| author_facet | McTeer, M Applegate, D Mesenbrink, P Ratziu, V Schattenberg, JM Bugianesi, E Geier, A Romero Gomez, M Dufour, J Ekstedt, M Francque, S Yki-Jarvinen, H Allison, M Valenti, L Miele, L Pavlides, M Cobbold, J Papatheodoridis, G Holleboom, AG Tiniakos, D Brass, C Anstee, QM Missier, P |
| author_sort | McTeer, M |
| collection | OXFORD |
| description | Aims: Metabolic dysfunction Associated Steatotic Liver Disease (MASLD) outcomes such as MASH (metabolic dysfunction associated steatohepatitis), fibrosis and cirrhosis are ordinarily determined by resource-intensive and invasive biopsies. We aim to show that routine clinical tests offer sufficient information to predict these endpoints. Methods: Using the LITMUS Metacohort derived from the European NAFLD Registry, the largest MASLD dataset in Europe, we create three combinations of features which vary in degree of procurement including a 19-variable feature set that are attained through a routine clinical appointment or blood test. This data was used to train predictive models using supervised machine learning (ML) algorithm XGBoost, alongside missing imputation technique MICE and class balancing algorithm SMOTE. Shapley Additive exPlanations (SHAP) were added to determine relative importance for each clinical variable. Results: Analysing nine biopsy-derived MASLD outcomes of cohort size ranging between 5385 and 6673 subjects, we were able to predict individuals at training set AUCs ranging from 0.719-0.994, including classifying individuals who are At-Risk MASH at an AUC = 0.899. Using two further feature combinations of 26-variables and 35-variables, which included composite scores known to be good indicators for MASLD endpoints and advanced specialist tests, we found predictive performance did not sufficiently improve. We are also able to present local and global explanations for each ML model, offering clinicians interpretability without the expense of worsening predictive performance. Conclusions: This study developed a series of ML models of accuracy ranging from 71.9—99.4% using only easily extractable and readily available information in predicting MASLD outcomes which are usually determined through highly invasive means. |
| first_indexed | 2024-09-25T04:05:18Z |
| format | Journal article |
| id | oxford-uuid:89e27393-16ab-4a96-95be-f2b5efeb18bf |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:05:18Z |
| publishDate | 2024 |
| publisher | Public Library of Science |
| record_format | dspace |
| spelling | oxford-uuid:89e27393-16ab-4a96-95be-f2b5efeb18bf2024-05-30T09:22:08ZMachine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical informationJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:89e27393-16ab-4a96-95be-f2b5efeb18bfEnglishJisc Publications RouterPublic Library of Science2024McTeer, MApplegate, DMesenbrink, PRatziu, VSchattenberg, JMBugianesi, EGeier, ARomero Gomez, MDufour, JEkstedt, MFrancque, SYki-Jarvinen, HAllison, MValenti, LMiele, LPavlides, MCobbold, JPapatheodoridis, GHolleboom, AGTiniakos, DBrass, CAnstee, QMMissier, PAims: Metabolic dysfunction Associated Steatotic Liver Disease (MASLD) outcomes such as MASH (metabolic dysfunction associated steatohepatitis), fibrosis and cirrhosis are ordinarily determined by resource-intensive and invasive biopsies. We aim to show that routine clinical tests offer sufficient information to predict these endpoints. Methods: Using the LITMUS Metacohort derived from the European NAFLD Registry, the largest MASLD dataset in Europe, we create three combinations of features which vary in degree of procurement including a 19-variable feature set that are attained through a routine clinical appointment or blood test. This data was used to train predictive models using supervised machine learning (ML) algorithm XGBoost, alongside missing imputation technique MICE and class balancing algorithm SMOTE. Shapley Additive exPlanations (SHAP) were added to determine relative importance for each clinical variable. Results: Analysing nine biopsy-derived MASLD outcomes of cohort size ranging between 5385 and 6673 subjects, we were able to predict individuals at training set AUCs ranging from 0.719-0.994, including classifying individuals who are At-Risk MASH at an AUC = 0.899. Using two further feature combinations of 26-variables and 35-variables, which included composite scores known to be good indicators for MASLD endpoints and advanced specialist tests, we found predictive performance did not sufficiently improve. We are also able to present local and global explanations for each ML model, offering clinicians interpretability without the expense of worsening predictive performance. Conclusions: This study developed a series of ML models of accuracy ranging from 71.9—99.4% using only easily extractable and readily available information in predicting MASLD outcomes which are usually determined through highly invasive means. |
| spellingShingle | McTeer, M Applegate, D Mesenbrink, P Ratziu, V Schattenberg, JM Bugianesi, E Geier, A Romero Gomez, M Dufour, J Ekstedt, M Francque, S Yki-Jarvinen, H Allison, M Valenti, L Miele, L Pavlides, M Cobbold, J Papatheodoridis, G Holleboom, AG Tiniakos, D Brass, C Anstee, QM Missier, P Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title | Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title_full | Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title_fullStr | Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title_full_unstemmed | Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title_short | Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information |
| title_sort | machine learning approaches to enhance diagnosis and staging of patients with masld using routinely available clinical information |
| work_keys_str_mv | AT mcteerm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT applegated machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT mesenbrinkp machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT ratziuv machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT schattenbergjm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT bugianesie machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT geiera machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT romerogomezm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT dufourj machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT ekstedtm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT francques machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT ykijarvinenh machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT allisonm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT valentil machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT mielel machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT pavlidesm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT cobboldj machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT papatheodoridisg machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT holleboomag machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT tiniakosd machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT brassc machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT ansteeqm machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT missierp machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation AT machinelearningapproachestoenhancediagnosisandstagingofpatientswithmasldusingroutinelyavailableclinicalinformation |