Systematic review and meta-analysis regarding impact of cataract surgery on biometric iris recognition
Mon statut pour la session
Quoi:
Paper Presentation | Présentation d'article
Quand:
5:07 PM, Samedi 2 Juin 2018
(3 minutes)
Où:
Metro Toronto Convention Centre (Édifice Sud)
- Salle 801
Authors: Daniel Stojanovic, Amit Sharma, Tony Lin
Author Disclosure Block: D. Stojanovic: None. A. Sharma: None. T. Lin: None.
Abstract Body:
Purpose: Biometrics has been increasingly used in security technologies to verify identities. Iris recognition in particular has become more popular around the globe, and has recently been integrated into the Canadian Nexus card program. This project aims to explore the impact of intra-ocular interventions, particularly cataract surgery, on the reliability of iris recognition.
Study Design: A systematic review and meta-analysis were conducted to determine the impact of cataract surgery on the rate of false-negatives when analyzed through iris recognition software. An electronic literature search of MEDLINE, EMBASE, and CINAHL was performed. Inclusion criteria consisted of pre- and post-surgical comparison of iris recognition.
Methods: In total, 462 articles were retrieved for primary analysis. For study inclusion, the exposure of interest was cataract surgery, and the primary outcome of interest was the false negative rate of iris recognition software when comparing pre-surgical to post-surgical eyes. In total, 4 studies were identified for inclusion.
Results: 235 eyes were analyzed. Each included study was a single armed trial comparing pre- and post- cataract surgery. A meta-analysis of proportions was used to aggregate the data. 6.723% (CI: 3.896 to 10.678) of participants received a false negative reading.
Conclusions: There is a risk of increased rates of false-negatives post cataract surgery. Consistently, this has been attributed to peri-operative damage to the iris. Under slit-lamp exam irises are successfully graded to indicate an increased risk of false-negative scans. It is unlikely that the existence of an intra-ocular lens within the eye has any intrinsic effect on biometric iris matching. However, patients may need documentation following cataract surgery to update their registered biometric data should any surgical complications occur.
Author Disclosure Block: D. Stojanovic: None. A. Sharma: None. T. Lin: None.
Abstract Body:
Purpose: Biometrics has been increasingly used in security technologies to verify identities. Iris recognition in particular has become more popular around the globe, and has recently been integrated into the Canadian Nexus card program. This project aims to explore the impact of intra-ocular interventions, particularly cataract surgery, on the reliability of iris recognition.
Study Design: A systematic review and meta-analysis were conducted to determine the impact of cataract surgery on the rate of false-negatives when analyzed through iris recognition software. An electronic literature search of MEDLINE, EMBASE, and CINAHL was performed. Inclusion criteria consisted of pre- and post-surgical comparison of iris recognition.
Methods: In total, 462 articles were retrieved for primary analysis. For study inclusion, the exposure of interest was cataract surgery, and the primary outcome of interest was the false negative rate of iris recognition software when comparing pre-surgical to post-surgical eyes. In total, 4 studies were identified for inclusion.
Results: 235 eyes were analyzed. Each included study was a single armed trial comparing pre- and post- cataract surgery. A meta-analysis of proportions was used to aggregate the data. 6.723% (CI: 3.896 to 10.678) of participants received a false negative reading.
Conclusions: There is a risk of increased rates of false-negatives post cataract surgery. Consistently, this has been attributed to peri-operative damage to the iris. Under slit-lamp exam irises are successfully graded to indicate an increased risk of false-negative scans. It is unlikely that the existence of an intra-ocular lens within the eye has any intrinsic effect on biometric iris matching. However, patients may need documentation following cataract surgery to update their registered biometric data should any surgical complications occur.
