INVESTIGATION OF THE RELATIONSHIP BETWEEN SEPTORHINOPLASTY AND FACIAL RECOGNITION SYSTEMS
Küçük Resim Yok
Tarih
2025
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Istanbul Univ, Fac Medicine, Publ Off
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Objectives: To investigate the temporal verification performance of the facial recognition systems after septorhinoplasty. Materialand Method: The study population included male and female patients who underwent septorhinoplasty at our institution between January 2022 and December 2023. Pre-and postoperative photographs were taken at 1, 2, and 4 weeks using the same camera, under the same distance, and under the same lighting conditions. In this technique-agnostic study, the analysis focused on the overall effect of the procedure rather than the impact of specific surgical manoeuvres. The change over time (preoperative, postoperative weeks 1, 2, 4) was compared based on the mean distance values in the face recognition systems. Results: The evaluation was conducted on 119 patients, comprising 75 females and 44 males with a mean age of 26.9 +/- 7.34 years (range, 18-56 years). When the accuracy rates of the face recognition systems were evaluated, the highest performance rate was obtained with the Euclidean metric for the VGG-Face system (94.85%). Among the face extraction methods, the RetinaFace (99.40%) and Mtcnn (99.19%) methods had the highest accuracy rates with the Euclidean metric in the VGG-Face face recognition system. There was a significant correlation between the mean distance value (0.378) in the preoperative-postoperative 2ndweek evaluation (0-2) and the mean distance value (0.279) in the 2nd-4th week evaluation (r=0.747, p=0.004). Conclusions: The alteration of facial components and appearance following septorhinoplasty remains a challenge for postoperative biometric verification using current facial recognition technolo gies. Rhinologists should be aware of the relationship between septorhinoplasty and facial recognition systems.
Açıklama
Anahtar Kelimeler
Facial Recognition Technology, Septorhinoplasty, Biometric Identification, Rhinoplasty, Deep Learning
Kaynak
Journal of Istanbul Faculty of Medicine-Istanbul Tip Fakultesi Dergisi
WoS Q Değeri
Q4
Scopus Q Değeri
Q4
Cilt
88
Sayı
4
