Keywords
medical diagnostic
Osteosarcoma detection
Random Forest
SMOTE
How to Cite
Abstract
Osteosarcoma is an aggressive and highly malignant bone cancer primarily affecting adolescents and young adults, with males being more commonly affected. Although deep learning models such as YOLO (95.73% accuracy) and VGG19 (95.25% accuracy), have demonstrated effectiveness in osteosarcoma detection, their large model sizes and extensive computational requirements limit their feasibility in resource-constrained environments. This study proposes a lightweight AI approach that optimizes osteosarcoma detection while maintaining high diagnostic accuracy, leveraging machine learning models under 5MB, manually or semi-automatically extracted features, and SMOTE for data balancing. Experimental results show that Random Forest, SVM, and XGBoost achieve accuracies of 94.70%, 94.23%, and 94.39%, respectively, closely matching the performance of YOLO and VGG19 while maintaining computational efficiency. Furthermore, the inference time for SVM is under one second (0.97s), demonstrating the speed advantage of lightweight models. These findings highlight the potential of small-size (lightweight) machine learning models to deliver high diagnostic accuracy with minimal computational requirements, providing a scalable and practical solution for early osteosarcoma detection in resource-limited settings. By balancing simplicity, efficiency, and high performance, this study establishes a new benchmark for achieving state-of-the-art results with lightweight models and paving the way for improved healthcare accessibility in underserved regions.
References
F. Mahyudin, M. Edward, M. Hardian Basuki, Y. Abdul Bari, Y. Suwandani, and C. Author, “HOSPITAL SURABAYA ‘A RETROSPECTIVE STUDY,’” no. 1, 2018, [Online]. Available: http://journal.unair.ac.id/ORTHO@journal-orthopaedi-and-traumatology-surabaya-media-104.html
M. P. Link et al., “The Effect of Adjuvant Chemotherapy on Relapse-Free Survival in Patients with Osteosarcoma of the Extremity,” New England Journal of Medicine, vol. 314, no. 25, pp. 1600–1606, Jun. 1986, doi: 10.1056/NEJM198606193142502.
“Key Statistics for Osteosarcoma.” Accessed: Dec. 07, 2024. [Online]. Available: https://www.cancer.org/cancer/types/osteosarcoma/about/key-statistics.html
M. A. Smith, S. F. Altekruse, P. C. Adamson, G. H. Reaman, and N. L. Seibel, “Declining childhood and adolescent cancer mortality,” Cancer, vol. 120, no. 16, pp. 2497–2506, Aug. 2014, doi: 10.1002/cncr.28748.
R. B. Guerra et al., “COMPARATIVE ANALYSIS BETWEEN OSTEOSARCOMA AND EWING’S SARCOMA: EVALUATION OF THE TIME FROM ONSET OF SIGNS AND SYMPTOMS UNTIL DIAGNOSIS,” Clinics, vol. 61, no. 2, pp. 99–106, Apr. 2006, doi: 10.1590/S1807-59322006000200003.
K. L. Pan, W. H. Chan, and Y. Y. Chia, “Initial Symptoms and Delayed Diagnosis of Osteosarcoma around the Knee Joint,” Journal of Orthopaedic Surgery, vol. 18, no. 1, pp. 55–57, Apr. 2010, doi: 10.1177/230949901001800112.
A. F. Kamal, H. Widyawarman, K. Husodo, E. U. Hutagalung, W. Rajabto, and A. F. Kamal, “Clinical Outcome and Survival of Osteosarcoma Patients in Cipto Mangunkusumo Hospital: Limb Salvage Surgery versus Amputation.”
American Cancer Society, “What Is Osteosarcoma?” Accessed: Dec. 07, 2024. [Online]. Available: https://www.cancer.org/cancer/types/osteosarcoma/about/what-is-osteosarcoma.html
L. Cai, J. Gao, and D. Zhao, “A review of the application of deep learning in medical image classification and segmentation,” Ann Transl Med, vol. 8, no. 11, pp. 713–713, Jun. 2020, doi: 10.21037/atm.2020.02.44.
W. Wang et al., “Medical Image Classification Using Deep Learning,” 2020, pp. 33–51. doi: 10.1007/978-3-030-32606-7_3.
S. Tamuly, C. Jyotsna, and J. Amudha, “Deep Learning Model for Image Classification,” 2020, pp. 312–320. doi: 10.1007/978-3-030-37218-7_36.
G. Algan and I. Ulusoy, “Image classification with deep learning in the presence of noisy labels: A survey,” Knowl Based Syst, vol. 215, p. 106771, Mar. 2021, doi: 10.1016/j.knosys.2021.106771.
S. Li, W. Song, L. Fang, Y. Chen, P. Ghamisi, and J. A. Benediktsson, “Deep Learning for Hyperspectral Image Classification: An Overview,” IEEE Transactions on Geoscience and Remote Sensing, vol. 57, no. 9, pp. 6690–6709, Sep. 2019, doi: 10.1109/TGRS.2019.2907932.
D. M. Anisuzzaman, H. Barzekar, L. Tong, J. Luo, and Z. Yu, “A deep learning study on osteosarcoma detection from histological images,” Biomed Signal Process Control, vol. 69, Aug. 2021, doi: 10.1016/j.bspc.2021.102931.
Patrick Leavey, Anita Sengupta, Dinesh Rakheja, Ovidiu Daescu, Harish Babu Arunachalam, and Rashika Mishra, “Osteosarcoma-Tumor-Assessment.” Accessed: Dec. 07, 2024. [Online]. Available: https://www.cancerimagingarchive.net/collection/osteosarcoma-tumor-assessment/
S. Gawade, A. Bhansali, K. Patil, and D. Shaikh, “Application of the convolutional neural networks and supervised deep-learning methods for osteosarcoma bone cancer detection,” Healthcare
Analytics, vol. 3, Nov. 2023, doi: 10.1016/j.health.2023.100153.
J. Li et al., “Primary bone tumor detection and classification in full-field bone radiographs via YOLO deep learning model”, doi: 10.1007/s00330-022-09289-y/Published.
Keras.io, “Keras Applications.” Accessed: Dec. 09, 2024. [Online]. Available: https://keras.io/api/applications/
W. Chen, K. Yang, Z. Yu, Y. Shi, and C. L. P. Chen, “A survey on imbalanced learning: latest research, applications and future directions,” Artif Intell Rev, vol. 57, no. 6, p. 137, May 2024, doi: 10.1007/s10462-024-10759-6.
M. T. Aziz et al., “A Novel Hybrid Approach for Classifying Osteosarcoma Using Deep Feature Extraction and Multilayer Perceptron,” Diagnostics, vol. 13, no. 12, Jun. 2023, doi: 10.3390/diagnostics13122106.
M. A. A. Walid et al., “Adapted Deep Ensemble Learning-Based Voting Classifier for Osteosarcoma Cancer Classification,” Diagnostics, vol. 13, no. 19, Oct. 2023, doi: 10.3390/diagnostics13193155.
H. Tang, H. Huang, J. Liu, J. Zhu, F. Gou, and J. Wu, “AI-Assisted Diagnosis and Decision-Making Method in Developing Countries for Osteosarcoma,” Healthcare (Switzerland), vol. 10, no. 11, Nov. 2022, doi: 10.3390/healthcare10112313.
T. Ouyang, S. Yang, F. Gou, Z. Dai, and J. Wu, “Rethinking U-Net from an Attention Perspective with Transformers for Osteosarcoma MRI Image Segmentation,” Comput Intell Neurosci, vol. 2022, 2022, doi: 10.1155/2022/7973404.
B. Lv, F. Liu, Y. Li, J. Nie, F. Gou, and J. Wu, “Artificial Intelligence-Aided Diagnosis Solution by Enhancing the Edge Features of Medical Images,” Diagnostics, vol. 13, no. 6, Mar. 2023, doi: 10.3390/diagnostics13061063.
J. Wu, Y. Guo, F. Gou, and Z. Dai, “A medical assistant segmentation method for MRI images of osteosarcoma based on DecoupleSegNet,” International Journal of Intelligent Systems, vol. 37, no. 11, pp. 8436–8461, Nov. 2022, doi: 10.1002/int.22949.
F. Liu, L. Xing, X. Zhang, and X. Zhang, “A four-pseudogene classifier identified by machine learning serves as a novel prognostic marker for survival of osteosarcoma,” Genes (Basel), vol. 10, no. 6, Jun. 2019, doi: 10.3390/genes10060414.
A. Fernandez, S. Garcia, F. Herrera, and N. V. Chawla, “SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary,” Journal of Artificial Intelligence Research, vol. 61, pp. 863–905, Apr. 2018, doi: 10.1613/jair.1.11192.
M. Mukherjee and M. Khushi, “SMOTE-ENC: A Novel SMOTE-Based Method to Generate Synthetic Data for Nominal and Continuous Features,” Applied System Innovation, vol. 4, no. 1, p. 18, Mar. 2021, doi: 10.3390/asi4010018.
D. Dablain, B. Krawczyk, and N. V. Chawla, “DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data,” IEEE Trans Neural Netw Learn Syst, vol. 34, no. 9, pp. 6390–6404, Sep. 2023, doi: 10.1109/TNNLS.2021.3136503.
R. Blagus and L. Lusa, “SMOTE for high-dimensional class-imbalanced data,” BMC Bioinformatics, vol. 14, no. 1, p. 106, Dec. 2013, doi: 10.1186/1471-2105-14-106.
N. V Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, “SMOTE: Synthetic Minority Over-sampling Technique,” 2002.
R. Oktafiani and E. Itje Sela, “Breast Cancer Classification with Principal Component Analysis and Smote using Random Forest Method and Support Vector Machine,” 2024. [Online]. Available: https://archive.ics.uci.edu/
Z. Zhao and T. Bai, “Financial Fraud Detection and Prediction in Listed Companies Using SMOTE and Machine Learning Algorithms,” Entropy, vol. 24, no. 8, Aug. 2022, doi: 10.3390/e24081157.
J. Wen, X. Tang, and J. Lu, “An imbalanced learning method based on graph tran-smote for fraud detection,” Sci Rep, vol. 14, no. 1, p. 16560, Jul. 2024, doi: 10.1038/s41598-024-67550-4.
E. Ileberi, Y. Sun, and Z. Wang, “Performance Evaluation of Machine Learning Methods for Credit Card Fraud Detection Using SMOTE and AdaBoost,” IEEE Access, vol. 9, pp. 165286–165294, 2021, doi: 10.1109/ACCESS.2021.3134330.
S. Hasan Abdulla, A. M. Sagheer, and H. Veisi, “Improving Breast Cancer Classification Using (Smote) Technique and Pectoral Muscle Removal in Mammographic Images,” MENDEL-Soft Computing Journal, vol. 2, pp. 2571–3701, 2021, doi: 10.13164/mendel.2021..0.
L. Huang, W. Xia, B. Zhang, B. Qiu, and X. Gao, “MSFCN-multiple supervised fully convolutional networks for the osteosarcoma segmentation of CT images,” Comput Methods Programs Biomed, vol. 143, pp. 67–74, May 2017, doi: 10.1016/j.cmpb.2017.02.013.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.