FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs
Background & purpose: For the best possible outcomes from therapy, proximal femur bone cancers must be accurately classified. This work creates an artificial intelligence (AI) model based on plain radiographs to categorize bone tumor in the proximal femur. Materials and methods: A tertiary r...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-10-01
|
| Series: | Journal of Bone Oncology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2212137423000374 |
| _version_ | 1797660309229404160 |
|---|---|
| author | Canyu Pan Luoyu Lian Jieyun Chen Risheng Huang |
| author_facet | Canyu Pan Luoyu Lian Jieyun Chen Risheng Huang |
| author_sort | Canyu Pan |
| collection | DOAJ |
| description | Background & purpose: For the best possible outcomes from therapy, proximal femur bone cancers must be accurately classified. This work creates an artificial intelligence (AI) model based on plain radiographs to categorize bone tumor in the proximal femur. Materials and methods: A tertiary referral center's standard anteroposterior hip radiographs were employed. A dataset 538 images of the femur, including malignant, benign, and tumor-free cases, was employed for training the AI model. There is a total of 214 images showing bone tumor. Pre-processing techniques were applied, and DenseNet model utilized for classification. The performance of the DenseNet model was compared to that of human doctors using cross-validation, further enhanced by incorporating Grad-CAM to visually indicate tumor locations. Results: For the three-label classification job, the suggested method boasts an excellent area under the receiver operating characteristic (AUROC) of 0.953. It scored much higher (0.853) than the diagnosis accuracy of the human experts in manual classification (0.794). The AI model outperformed the mean values of the clinicians in terms of sensitivity, specificity, accuracy, and F1 scores. Conclusion: The developed DenseNet model demonstrated remarkable accuracy in classifying bone tumors in the proximal femur using plain radiographs. This technology has the potential to reduce misdiagnosis, particularly among non-specialists in musculoskeletal oncology. The utilization of advanced deep learning models provides a promising approach for improved classification and enhanced clinical decision-making in bone tumor detection. |
| first_indexed | 2024-03-11T18:28:56Z |
| format | Article |
| id | doaj.art-70c9a0d834a54f0c836d17d2b7f89db0 |
| institution | Directory Open Access Journal |
| issn | 2212-1374 |
| language | English |
| last_indexed | 2024-03-11T18:28:56Z |
| publishDate | 2023-10-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Bone Oncology |
| spelling | doaj.art-70c9a0d834a54f0c836d17d2b7f89db02023-10-13T13:53:40ZengElsevierJournal of Bone Oncology2212-13742023-10-0142100504FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographsCanyu Pan0Luoyu Lian1Jieyun Chen2Risheng Huang3Department of Radiology, Quanzhou First Hospital Affiliated to Fujian Medical University, Quanzhou 362000, Fujian Province, ChinaDepartment of Thoracic Surgery, Quanzhou First Hospital Affiliated to Fujian Medical, University, Quanzhou 362000, Fujian Province, ChinaDepartment of Radiology, Quanzhou First Hospital Affiliated to Fujian Medical University, Quanzhou 362000, Fujian Province, China; Corresponding author at: Department of Radiology, Quanzhou First Hospital Affiliated to Fujian Medical University, 248-252 East Street, Licheng District, Quanzhou City, Fujian Province 362000, China.Department of Radiology, Quanzhou First Hospital Affiliated to Fujian Medical University, Quanzhou 362000, Fujian Province, ChinaBackground & purpose: For the best possible outcomes from therapy, proximal femur bone cancers must be accurately classified. This work creates an artificial intelligence (AI) model based on plain radiographs to categorize bone tumor in the proximal femur. Materials and methods: A tertiary referral center's standard anteroposterior hip radiographs were employed. A dataset 538 images of the femur, including malignant, benign, and tumor-free cases, was employed for training the AI model. There is a total of 214 images showing bone tumor. Pre-processing techniques were applied, and DenseNet model utilized for classification. The performance of the DenseNet model was compared to that of human doctors using cross-validation, further enhanced by incorporating Grad-CAM to visually indicate tumor locations. Results: For the three-label classification job, the suggested method boasts an excellent area under the receiver operating characteristic (AUROC) of 0.953. It scored much higher (0.853) than the diagnosis accuracy of the human experts in manual classification (0.794). The AI model outperformed the mean values of the clinicians in terms of sensitivity, specificity, accuracy, and F1 scores. Conclusion: The developed DenseNet model demonstrated remarkable accuracy in classifying bone tumors in the proximal femur using plain radiographs. This technology has the potential to reduce misdiagnosis, particularly among non-specialists in musculoskeletal oncology. The utilization of advanced deep learning models provides a promising approach for improved classification and enhanced clinical decision-making in bone tumor detection.http://www.sciencedirect.com/science/article/pii/S2212137423000374Bone tumorsProximal femurArtificial intelligenceRadiographsDenseNetClassification |
| spellingShingle | Canyu Pan Luoyu Lian Jieyun Chen Risheng Huang FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs Journal of Bone Oncology Bone tumors Proximal femur Artificial intelligence Radiographs DenseNet Classification |
| title | FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs |
| title_full | FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs |
| title_fullStr | FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs |
| title_full_unstemmed | FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs |
| title_short | FemurTumorNet: Bone tumor classification in the proximal femur using DenseNet model based on radiographs |
| title_sort | femurtumornet bone tumor classification in the proximal femur using densenet model based on radiographs |
| topic | Bone tumors Proximal femur Artificial intelligence Radiographs DenseNet Classification |
| url | http://www.sciencedirect.com/science/article/pii/S2212137423000374 |
| work_keys_str_mv | AT canyupan femurtumornetbonetumorclassificationintheproximalfemurusingdensenetmodelbasedonradiographs AT luoyulian femurtumornetbonetumorclassificationintheproximalfemurusingdensenetmodelbasedonradiographs AT jieyunchen femurtumornetbonetumorclassificationintheproximalfemurusingdensenetmodelbasedonradiographs AT rishenghuang femurtumornetbonetumorclassificationintheproximalfemurusingdensenetmodelbasedonradiographs |