Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions

Abstract Objective To evaluate the diagnostic performance of a machine learning model combining deep learning models based on enhanced CT images with radiological and clinical features in differentiating lipid-poor adrenal adenomas from metastatic tumors, and to explain the model’s prediction result...

Full description

Saved in:
Bibliographic Details
Main Authors: Shao-Cai Wang, Sheng-Nan Yin, Zi-You Wang, Ning Ding, Yi-Ding Ji, Long Jin
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Medical Imaging
Subjects:
Online Access:https://doi.org/10.1186/s12880-025-01798-8
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849334003199377408
author Shao-Cai Wang
Sheng-Nan Yin
Zi-You Wang
Ning Ding
Yi-Ding Ji
Long Jin
author_facet Shao-Cai Wang
Sheng-Nan Yin
Zi-You Wang
Ning Ding
Yi-Ding Ji
Long Jin
author_sort Shao-Cai Wang
collection DOAJ
description Abstract Objective To evaluate the diagnostic performance of a machine learning model combining deep learning models based on enhanced CT images with radiological and clinical features in differentiating lipid-poor adrenal adenomas from metastatic tumors, and to explain the model’s prediction results through SHAP(Shapley Additive Explanations) analysis. Methods A retrospective analysis was conducted on abdominal contrast-enhanced CT images and clinical data from 416 pathologically confirmed adrenal tumor patients at our hospital from July 2019 to December 2024. Patients were randomly divided into training and testing sets in a 7:3 ratio. Six convolutional neural network (CNN)-based deep learning models were employed, and the model with the highest diagnostic performance was selected based on the area under curve(AUC) of the ROC. Subsequently, multiple machine learning models incorporating clinical and radiological features were developed and evaluated using various indicators and AUC.The best-performing machine learning model was further analyzed using SHAP plots to enhance interpretability and quantify feature contributions. Results All six deep learning models demonstrated excellent diagnostic performance, with AUC values exceeding 0.8, among which ResNet50 achieved the highest AUC. Among the 10 machine learning models incorporating clinical and imaging features, the extreme gradient boosting(XGBoost) model exhibited the best accuracy(ACC), sensitivity, and AUC, indicating superior diagnostic performance.SHAP analysis revealed contributions from ResNet50, RPW, age, and other key features in model predictions. Conclusion Machine learning models based on contrast-enhanced CT combined with clinical and imaging features exhibit outstanding diagnostic performance in differentiating lipid-poor adrenal adenomas from metastases.
format Article
id doaj-art-6edf7c0177c4473eaed5d9e2c4cab1f2
institution Kabale University
issn 1471-2342
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Medical Imaging
spelling doaj-art-6edf7c0177c4473eaed5d9e2c4cab1f22025-08-20T03:45:41ZengBMCBMC Medical Imaging1471-23422025-07-012511910.1186/s12880-025-01798-8Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesionsShao-Cai Wang0Sheng-Nan Yin1Zi-You Wang2Ning Ding3Yi-Ding Ji4Long Jin5Suzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversitySuzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversitySuzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversitySuzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversitySuzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversitySuzhou Ninth People’s Hospital, Suzhou Ninth Hospital Affiliated to Soochow UniversityAbstract Objective To evaluate the diagnostic performance of a machine learning model combining deep learning models based on enhanced CT images with radiological and clinical features in differentiating lipid-poor adrenal adenomas from metastatic tumors, and to explain the model’s prediction results through SHAP(Shapley Additive Explanations) analysis. Methods A retrospective analysis was conducted on abdominal contrast-enhanced CT images and clinical data from 416 pathologically confirmed adrenal tumor patients at our hospital from July 2019 to December 2024. Patients were randomly divided into training and testing sets in a 7:3 ratio. Six convolutional neural network (CNN)-based deep learning models were employed, and the model with the highest diagnostic performance was selected based on the area under curve(AUC) of the ROC. Subsequently, multiple machine learning models incorporating clinical and radiological features were developed and evaluated using various indicators and AUC.The best-performing machine learning model was further analyzed using SHAP plots to enhance interpretability and quantify feature contributions. Results All six deep learning models demonstrated excellent diagnostic performance, with AUC values exceeding 0.8, among which ResNet50 achieved the highest AUC. Among the 10 machine learning models incorporating clinical and imaging features, the extreme gradient boosting(XGBoost) model exhibited the best accuracy(ACC), sensitivity, and AUC, indicating superior diagnostic performance.SHAP analysis revealed contributions from ResNet50, RPW, age, and other key features in model predictions. Conclusion Machine learning models based on contrast-enhanced CT combined with clinical and imaging features exhibit outstanding diagnostic performance in differentiating lipid-poor adrenal adenomas from metastases.https://doi.org/10.1186/s12880-025-01798-8Adrenal glandAdenomaMetastasisDeep learningMachine learningAutomatic segmentation
spellingShingle Shao-Cai Wang
Sheng-Nan Yin
Zi-You Wang
Ning Ding
Yi-Ding Ji
Long Jin
Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
BMC Medical Imaging
Adrenal gland
Adenoma
Metastasis
Deep learning
Machine learning
Automatic segmentation
title Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
title_full Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
title_fullStr Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
title_full_unstemmed Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
title_short Evaluation of a fusion model combining deep learning models based on enhanced CT images with radiological and clinical features in distinguishing lipid-poor adrenal adenoma from metastatic lesions
title_sort evaluation of a fusion model combining deep learning models based on enhanced ct images with radiological and clinical features in distinguishing lipid poor adrenal adenoma from metastatic lesions
topic Adrenal gland
Adenoma
Metastasis
Deep learning
Machine learning
Automatic segmentation
url https://doi.org/10.1186/s12880-025-01798-8
work_keys_str_mv AT shaocaiwang evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions
AT shengnanyin evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions
AT ziyouwang evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions
AT ningding evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions
AT yidingji evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions
AT longjin evaluationofafusionmodelcombiningdeeplearningmodelsbasedonenhancedctimageswithradiologicalandclinicalfeaturesindistinguishinglipidpooradrenaladenomafrommetastaticlesions