A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107.
In this study, we developed a convolutional neural network approach for directly classifying digital imaging and communication in medicine files in medical imaging applications. Existing models require converting this format into other formats like portable network graphics. This conversion leads to...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
wiley
2025
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/20.500.12493/2925 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841524063496830976 |
|---|---|
| author | Mabirizi, Vicent Wasswa, William Kawuma, Simon |
| author_facet | Mabirizi, Vicent Wasswa, William Kawuma, Simon |
| author_sort | Mabirizi, Vicent |
| collection | KAB-DR |
| description | In this study, we developed a convolutional neural network approach for directly classifying digital imaging and communication in medicine files in medical imaging applications. Existing models require converting this format into other formats like portable network graphics. This conversion leads to metadata loss and classification bias, the developed model processes raw digital imaging and communication in medicine files, thereby preserving both pixel data and embedded metadata. The model was evaluated on chest X-ray images for tuberculosis detection and magnetic resonance imaging scan images for brain tumour classification from the National Institute of Allergy and Infectious Diseases. The X-ray modality achieved a precision of 92.9%, recall of 88.4%, F1-score of 90.6% and accuracy of 90.9%, while the magnetic resonance imaging modality obtained a precision of 80.0%, recall of 79.4%, F1-score of 79.7% and accuracy of 85.5%. These results demonstrate the model’s effectiveness across multiple imaging modalities. A key advantage of this approach is the preservation of diagnostic metadata, enhancing accuracy and reducing classification bias. The study highlights its potential to improve medical imaging and support real-time clinical decision making. Despite the promising results, the study acknowledges limitations in dataset diversity and computational efficiency, with future work focusing on addressing these challenges and further optimising the model for deployment in resource-limited environments. |
| format | Article |
| id | oai:idr.kab.ac.ug:20.500.12493-2925 |
| institution | KAB-DR |
| language | English |
| publishDate | 2025 |
| publisher | wiley |
| record_format | dspace |
| spelling | oai:idr.kab.ac.ug:20.500.12493-29252025-07-19T00:00:28Z A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. Mabirizi, Vicent Wasswa, William Kawuma, Simon Multimodal convolutional neural network (CNN) DICOM file classification Raw DICOM processing Metadata preservation Image pixel data and metadata fusion In this study, we developed a convolutional neural network approach for directly classifying digital imaging and communication in medicine files in medical imaging applications. Existing models require converting this format into other formats like portable network graphics. This conversion leads to metadata loss and classification bias, the developed model processes raw digital imaging and communication in medicine files, thereby preserving both pixel data and embedded metadata. The model was evaluated on chest X-ray images for tuberculosis detection and magnetic resonance imaging scan images for brain tumour classification from the National Institute of Allergy and Infectious Diseases. The X-ray modality achieved a precision of 92.9%, recall of 88.4%, F1-score of 90.6% and accuracy of 90.9%, while the magnetic resonance imaging modality obtained a precision of 80.0%, recall of 79.4%, F1-score of 79.7% and accuracy of 85.5%. These results demonstrate the model’s effectiveness across multiple imaging modalities. A key advantage of this approach is the preservation of diagnostic metadata, enhancing accuracy and reducing classification bias. The study highlights its potential to improve medical imaging and support real-time clinical decision making. Despite the promising results, the study acknowledges limitations in dataset diversity and computational efficiency, with future work focusing on addressing these challenges and further optimising the model for deployment in resource-limited environments. 2025-07-18T08:34:14Z 2025-07-18T08:34:14Z 2025 Article Vicent, M., Willian, W., & Simon, K. (2025). A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. The Journal of Engineering, 2025(1), e70107. https://doi.org/10.1049/tje2.70107 http://hdl.handle.net/20.500.12493/2925 en Attribution 3.0 United States http://creativecommons.org/licenses/by/3.0/us/ application/pdf wiley |
| spellingShingle | Multimodal convolutional neural network (CNN) DICOM file classification Raw DICOM processing Metadata preservation Image pixel data and metadata fusion Mabirizi, Vicent Wasswa, William Kawuma, Simon A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title | A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title_full | A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title_fullStr | A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title_full_unstemmed | A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title_short | A Multimodal Convolutional Neural Network Based Approach for DICOM Files Classification. , 2025(1), e70107. |
| title_sort | multimodal convolutional neural network based approach for dicom files classification 2025 1 e70107 |
| topic | Multimodal convolutional neural network (CNN) DICOM file classification Raw DICOM processing Metadata preservation Image pixel data and metadata fusion |
| url | http://hdl.handle.net/20.500.12493/2925 |
| work_keys_str_mv | AT mabirizivicent amultimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 AT wasswawilliam amultimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 AT kawumasimon amultimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 AT mabirizivicent multimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 AT wasswawilliam multimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 AT kawumasimon multimodalconvolutionalneuralnetworkbasedapproachfordicomfilesclassification20251e70107 |