Explainable machine learning model based on clinical factors for predicting the disappearance of indeterminate pulmonary nodules.
Clinical factor
Explainable machine learning
Feature importance
Indeterminate pulmonary nodule
Visualization
Journal
Computers in biology and medicine
ISSN: 1879-0534
Titre abrégé: Comput Biol Med
Pays: United States
ID NLM: 1250250
Informations de publication
Date de publication:
22 Dec 2023
22 Dec 2023
Historique:
received:
25
07
2023
revised:
01
11
2023
accepted:
17
12
2023
medline:
29
12
2023
pubmed:
29
12
2023
entrez:
28
12
2023
Statut:
aheadofprint
Résumé
During lung cancer screening, indeterminate pulmonary nodules (IPNs) are a frequent finding. We aim to predict whether IPNs are resolving or non-resolving to reduce follow-up examinations, using machine learning (ML) models. We incorporated dedicated techniques to enhance prediction explainability. In total, 724 IPNs (size 50-500 mm The random forest model outperformed the other ML models with an AUC of 0.865. This model achieved a recall of 0.646, a precision of 0.816, and an F1 score of 0.721. The evaluation of feature importance achieved consistent ranking across all three methods for the most crucial factors. The MDI, PFI, and SHAP methods highlighted volume, maximum diameter, and minimum diameter as the top three factors. However, the remaining factors revealed discrepant ranking across methods. ML models effectively predict IPN disappearance using participant demographics and nodule characteristics. Explainable techniques can assist clinicians in developing understandable preliminary assessments.
Sections du résumé
BACKGROUND
BACKGROUND
During lung cancer screening, indeterminate pulmonary nodules (IPNs) are a frequent finding. We aim to predict whether IPNs are resolving or non-resolving to reduce follow-up examinations, using machine learning (ML) models. We incorporated dedicated techniques to enhance prediction explainability.
METHODS
METHODS
In total, 724 IPNs (size 50-500 mm
RESULTS
RESULTS
The random forest model outperformed the other ML models with an AUC of 0.865. This model achieved a recall of 0.646, a precision of 0.816, and an F1 score of 0.721. The evaluation of feature importance achieved consistent ranking across all three methods for the most crucial factors. The MDI, PFI, and SHAP methods highlighted volume, maximum diameter, and minimum diameter as the top three factors. However, the remaining factors revealed discrepant ranking across methods.
CONCLUSION
CONCLUSIONS
ML models effectively predict IPN disappearance using participant demographics and nodule characteristics. Explainable techniques can assist clinicians in developing understandable preliminary assessments.
Identifiants
pubmed: 38154157
pii: S0010-4825(23)01336-7
doi: 10.1016/j.compbiomed.2023.107871
pii:
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
107871Informations de copyright
Copyright © 2023 The Authors. Published by Elsevier Ltd.. All rights reserved.
Déclaration de conflit d'intérêts
Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.