GradCAM as an explicability method to evaluate the performance of deep learning models in classifying pediatric arteriovenous malformations (AVM) in arterial spin labeling sequences (ASL)
Por:
Romagosa J, Mata-Miquel C, Benítez R, Valls-Esteve A, Bernaus S, Ibnoulkhatib M, Stephan-Otto C and Munuera J
Publicada:
20 oct 2025
Ahead of Print:
20 oct 2025
Resumen:
PurposeThe study investigates the usefulness of Convolutional Neural Networks (CNNs) in accurately detecting arteriovenous malformations in pediatric medical imaging, particularly using arterial spin labeling sequences. It also aims to offer diagnostic explanations comparable to expert analysis.MethodsThe research analyzed three different CNN architectures to determine their performance in detecting arteriovenous malformations. The study focused on evaluating the relationship between model complexity and performance increase, using data to assess the accuracy and diagnostic usefulness of each model.ResultsThe findings indicated a nonlinear link between model complexity and performance. Sur- prisingly, more complex models frequently produced poor results and diagnostically useless answers. The simplest CNN models achieved the highest accuracy rate (90%), demonstrating the effectiveness of minimal complexity in model construction. Heat maps showed a strong association with the real locations of irregularities, indicating that the models were interpretable.ConclusionThe study highlights the usefulness of CNNs in medical diagnostics, emphasizing the importance of model simplicity and interpretability in clinical applications. It suggests a need for balancing technical sophistication with clinical value and presents options for future research into refining CNN structures for increased diagnostic precision in various medical imaging modalities.
Filiaciones:
Romagosa J:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Mata-Miquel C:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Institute for Research and Innovation in Health (IRIS), BIOCOM-SC, Universitat Politècnica de Catalunya, Barcelona, Spain. ROR: https://ror.org/03mb6wj31. GRID: grid.6835.8. ISNI: 0000 0004 1937 028X
Benítez R:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Institute for Research and Innovation in Health (IRIS), BIOCOM-SC, Universitat Politècnica de Catalunya, Barcelona, Spain. ROR: https://ror.org/03mb6wj31. GRID: grid.6835.8. ISNI: 0000 0004 1937 028X
Valls-Esteve A:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Innovation Department, Hospital Sant Joan de Déu, Esplugues del Llobregat, Spain. ROR: https://ror.org/001jx2139. GRID: grid.411160.3. ISNI: 0000 0001 0663 8628
Bernaus S:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Ibnoulkhatib M:
Department of Diagnostic Imaging, Hospital Sant Joan de Déu, Esplugues del Llobregat, Spain. ROR: https://ror.org/001jx2139. GRID: grid.411160.3. ISNI: 0000 0001 0663 8628
Stephan-Otto C:
Pediatric Computational Imaging Center, Institut de Recerca Sant Joan de Déu, Esplugues de Llobregat, Spain. ROR: https://ror.org/00gy2ar74. ISNI: 0000 0004 9332 2809
Centro de Investigacion Biomédica en Red de Salud Mental, Madrid, Spain. ROR: https://ror.org/009byq155. GRID: grid.469673.9. ISNI: 0000 0004 5901 7501
Munuera J:
Advanced Medical Imaging, Artificial Intelligence, and Imaging-Guided Therapy, Institut de Recerca Sant Pau, Barcelona, Spain. ROR: https://ror.org/005teat46
Diagnostic Imaging Department, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. ROR: https://ror.org/059n1d175. GRID: grid.413396.a. ISNI: 0000 0004 1768 8905
Green Accepted
|