TY - GEN
T1 - Mitigating Data Bias and Errors Through Machine Unlearning for Accurate Human Skin Disease Prediction
AU - Maram, Balajee
AU - Perla, Someswari
AU - Nagaraju, D.
AU - Sathishkumar, V. E.
AU - Ramya, Vaggi
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.
PY - 2026
Y1 - 2026
N2 - Skin disorders affect nearly 900 million individuals worldwide, but diagnostic error alone is responsible for nearly 20% of misdiagnosis and requires more accurate predictive systems to be in place. While artificial intelligence (AI) algorithms hold great potential for dermatalogic diagnosis, they are often undermined by erroneous, skewed, or wrongly labeled training datasets. Here, we introduce a selective erasure model that erases biased or corrupted training samples without full retraining. On a densely annotated dermatology image dataset of over 25,000 labeled images, our scheme enhanced diagnostic accuracy by 12.8%, reducing error rates substantially from conventional deep learning models. The method also minimized false positives by 21.4%, improving both clinical safety and credibility. Beyond performance enhancements, the methodology enables machine unlearning techniques that ensure patient anonymity, ethical application of AI, and adaptability, forming the foundation for secure and equitable skin disease prediction models.
AB - Skin disorders affect nearly 900 million individuals worldwide, but diagnostic error alone is responsible for nearly 20% of misdiagnosis and requires more accurate predictive systems to be in place. While artificial intelligence (AI) algorithms hold great potential for dermatalogic diagnosis, they are often undermined by erroneous, skewed, or wrongly labeled training datasets. Here, we introduce a selective erasure model that erases biased or corrupted training samples without full retraining. On a densely annotated dermatology image dataset of over 25,000 labeled images, our scheme enhanced diagnostic accuracy by 12.8%, reducing error rates substantially from conventional deep learning models. The method also minimized false positives by 21.4%, improving both clinical safety and credibility. Beyond performance enhancements, the methodology enables machine unlearning techniques that ensure patient anonymity, ethical application of AI, and adaptability, forming the foundation for secure and equitable skin disease prediction models.
KW - Data Bias Mitigation
KW - Ethical Healthcare
KW - Machine Unlearning
KW - Medical AI
KW - Skin Disease Prediction
UR - https://www.scopus.com/pages/publications/105032040083
U2 - 10.1007/978-3-032-17834-3_6
DO - 10.1007/978-3-032-17834-3_6
M3 - Conference contribution
AN - SCOPUS:105032040083
SN - 9783032178336
T3 - Communications in Computer and Information Science
SP - 77
EP - 87
BT - Artificial Intelligence Based Smart and Secured Applications - 4th International Conference, ASCIS 2025, Revised Selected Papers
A2 - Rajagopal, Sridaran
A2 - Sajja, Priti
A2 - Thanki, Rohit
A2 - Kumar, Ajay
PB - Springer Science and Business Media Deutschland GmbH
T2 - 4th International Conference on Advancements in Smart Computing and Information Security, ASCIS 2025
Y2 - 11 September 2025 through 13 September 2025
ER -