MobileNetV2 and EfficientNetB3 Ensemble for Scalable Dermoscopic Skin Lesion Classification.: Ensemble Model for Dermoscopic Skin Lesions

Muhammad  Faris; Ramsha  Qayyum; Mir Farooq  Ali; Muhammad Mansoor  Mughal; Tariq  Javid

Authors

Muhammad Faris Hamdard University, Karachi, Pakistan
Ramsha Qayyum Hamdard University, Karachi, Pakistan https://orcid.org/0009-0007-4237-4966
Mir Farooq Ali Marche Polytechnic University, Ancona, Italy https://orcid.org/0009-0007-1457-5736
Muhammad Mansoor Mughal Hamdard University, Karachi, Pakistan
Tariq Javid Hamdard University, Karachi, Pakistan

Keywords:

Deep Learning, Dermoscopic Images, Transfer Learning, Tele-dermatology, Skin Disease Classification

Abstract

Background: Skin diseases, ranging from mild benign lesions to life-threatening malignant conditions, remain a major global health concern. Early and accurate diagnosis is critical to avoid complications, yet this is often limited by the shortage of dermatologists and the subjective nature of visual inspections, particularly in low-resource settings. To address this challenge, this study proposes an automated deep learning framework for skin disease classification using dermoscopic images.

Methods: The framework employs a hybrid learning approach by integrating transfer learning and ensemble learning techniques. Specifically, MobileNetV2 and EfficientNetB3 models were combined to leverage their unique strengths, thereby enhancing generalization and predictive accuracy. The system was trained on a well-annotated dataset of 22,177 dermoscopic images, representing eight diagnostic categories that include benign, malignant, and pre-cancerous skin conditions.

Results: Experimental results demonstrated strong classification performance, achieving a training accuracy of 96.81%, validation accuracy of 87.66% (loss of 0.455), and test accuracy of 86%. To improve clinical trust and interpretability, Gradient-weighted Class Activation Mapping (Grad-CAM) was utilized to highlight the image regions that contributed most to the model’s decisions. In addition, a user-friendly diagnostic interface was developed, enabling real-time image input, automated analysis, and clear interpretive guidance. This makes the system accessible not only to healthcare providers but also to non-specialists, bridging gaps in dermatological care.

Conclusion: The proposed solution offers a reliable, interpretable, and scalable application of artificial intelligence for skin disease screening, with significant implications for tele-dermatology and seamless integration into clinical workflows.

DOI: https://doi.org/10.59564/amrj/03.04/014

Author Biographies

Muhammad Faris, Hamdard University, Karachi, Pakistan

Assistant Professor, Department of Biomedical Engineering

Ramsha Qayyum, Hamdard University, Karachi, Pakistan

Officer, VC Office

Mir Farooq Ali, Marche Polytechnic University, Ancona, Italy

PhD Scholar, Department of Information Engineering

Muhammad Mansoor Mughal, Hamdard University, Karachi, Pakistan

Sr. Lecturer/PhD Scholar, Dept. of Electrical & Comp. Engg

Tariq Javid, Hamdard University, Karachi, Pakistan

Professor, Department of Biomedical Engineering

References

Koufidis C, Manninen K, Nieminen J, Wohlin M, Silén C. Representation, interaction and interpretation. Making sense of the context in clinical reasoning. Med Educ. 2022 Jan;56(1):98-109.

DOI: https://doi.org/10.1111/medu.14545

Currens JA, Bithell CP. Clinical education: listening to different perspectives. Physiotherapy. 2000 Dec 1;86(12):645-53.

Masic I. E-learning as new method of medical education. Acta Inform Med. 2008;16(2):102-17.

DOI: https://doi.org/10.5455/aim.2008.16.102-117

Picciano, A.G., Dziuban, C.D., Graham, C.R., & Moskal, P.D. (Eds.). (2021). Blended Learning: Research Perspectives, Volume 3 (1st ed.). Routledge.

DOI: https://doi.org/10.4324/9781003037736

Fu XT, Hu Y, Yan BC, Jiao YG, Zheng SJ, Wang YG, et al. The Use of Blended Teaching in Higher Medical Education during the Pandemic Era. Int J Clin Pract. 2022 Nov 14;2022:3882975.

DOI: https://doi.org/10.1155/2022/3882975

Ashraf MA, Tsegay SM, Gull N, Saeed M, Dawood H. The role of blended learning in improving medical students' academic performance: evidence from Pakistan. Front Med. 2024;11:1425659.

DOI: https://doi.org/10.3389/fmed.2024.1425659

Khan UI, Farazdaq H, Naseem A, Suleman W, Saleem S, Qadir MA, et al. Evaluation of FamMed essentials: a blended-learning program for capacity building of general practitioners in Pakistan. BMC Med Educ. 2024 Mar 1;24(1):218.

DOI: https://doi.org/10.1186/s12909-024-05069-y

Asad MM, Hussain N, Wadho M, Khand ZH, Churi PP. Integration of e-learning technologies for interactive teaching and learning process: an empirical study on higher education institutes of Pakistan. Journal of Applied Research in Higher Education. 2020;13(3):649-63.

DOI: https://doi.org/10.1108/jarhe-04-2020-0103

Avazmatova M. Significance of blended learning in education system [Internet]. ResearchGate; 2020. Available from:

https://www.researchgate.net/publication/343970415_Significance_Of_Blended_Learning_In_Education_SysTem. https://doi.org/10.37547/tajssei/Volume02Issue08-82

Tong DH, Uyen BP, Ngan LK. The effectiveness of blended learning on students' academic achievement, self-study skills and learning attitudes: A quasi-experiment study in teaching the conventions for coordinates in the plane. Heliyon. 2022;8(12):e12657.

DOI: https://doi.org/10.1016/j.heliyon.2022.e12657

Elaoufy H. The Adoption of Blended Learning as an Emerging Mode of Learning among EFL Students: Current Challenges and Future Directions. International Journal of Current Science Research and Review [Internet]. 2023 May 8;06(05).

DOI: https://doi.org/10.47191/ijcsrr/v6-i5-

Khawaja RA, et al. Perceptions and Effectiveness of Faculty Development Programs in Medical Education: A Study at a Pakistani Medical College. Journal of Rawalpindi Medical College. 2022;26(3):425-30.

DOI: https://doi.org/10.37939/jrmc.v27i2.2176

Peeraer G, Van Petegem P. The Role of Teaching Portfolios in the Professional Development of Medical Educators: A Scoping Review. Medical Teacher. 2021;43(4):389-97.

DOI: https://doi.org/10.1080/0142159X.2021.1876834

Moore DE, Green JS, Gallis HA. Achieving Desired Results and Improved Outcomes: Integrating Planning and Assessment Throughout Learning Activities. The Journal of Continuing Education in the Health Professions. 2009;29(1):1-DOI: https://doi.org/10.1002/chp.20001

Kiran F, Javaid A, Irum S, Zahoor A, Farooq F. Factors affecting professional identity formation of basic medical sciences teachers in Pakistan: a phenomenological analysis of interviews. InFrontiers in Education 2024 Jun 3 (Vol. 9, p. 1307560). Frontiers Media SA.

DOI: https://doi.org/10.3389/feduc.2024.1307560

Liu Q, Peng W, Zhang F, Hu R, Li Y, Yan W. The effectiveness of blended learning in health professions: Systematic review and meta-analysis. Journal of Medical Internet Research. 2016;18(1):e2.

DOI: https://doi.org/10.2196/jmir.4807

Rowe M, Frantz J, Bozalek V. The role of blended learning in the clinical education of healthcare students: A systematic review. Medical Teacher. 2012;34(4):e216-21.

DOI: https://doi.org/10.3109/0142159x.2012.642831