• Mr. Ashok Kumar Reddy
• Mr.B.Rameswara Reddy

The association between the face and sickness has been debated for thousands of years, leading to the emergence of facial diagnosis. Deep learning algorithms will be used to investigate the possibility of diagnosing diseases from uncontrolled 2D facial photos. We propose leveraging deep transfer learning from face recognition to provide computer-aided facial diagnosis on several diseases in this research. In the trials, we use a small dataset to perform computer-aided face diagnosis on single and multiple disorders (beta-thalassemia, hyperthyroidism, Down syndrome, and leprosy). Deep transfer learning from face recognition can achieve top-1 accuracy of more than 90%, outperforming both classic machine learning approaches and doctors in the studies. In practice, gathering disease-specific face photos is difficult, expensive, and time intensive, and it involves ethical constraints due to the treatment of personal data. As a result, datasets for face diagnostic research are often private and tiny in comparison to those for other machine learning application fields. Deep transfer learning algorithms that succeed in facial diagnosis with a short dataset could provide a low-cost and non-invasive method for illness screening and detection.

Facial Recognition based Diagnosis Using Transfer Learning

"Facial Recognition based Diagnosis Using Transfer Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 7, page no.g664-g669, July-2023, Available :http://www.jetir.org/papers/JETIR2307688.pdf

"Facial Recognition based Diagnosis Using Transfer Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 7, page no. ppg664-g669, July-2023, Available at : http://www.jetir.org/papers/JETIR2307688.pdf