New findings suggest the potential of a new artificial intelligence-based system for autonomous monitoring of patients treated for neovascular age-related macular degeneration (nAMD).
Research has demonstrated both its safety and compliance with expert consensus as being on par with decisions made in clinical practice, with particularly low rates of false positive classification of choroidal neovascularization (CNV) activity.
“More than half of eyes with AMD could be monitored independently without additional risk to the patient,” wrote study author Ivan Potapenko, MD, PhD, Department of Ophthalmology, Rigshospitalet. “The current results are encouraging; however, a live implementation is required to establish real performance.
Management strategies for AMD generally require close monitoring and the administration of anti-VEGF injections for long periods of time. In combination with the demographic transition to an older population, an increased burden will affect ophthalmology services.
In order to maintain quality of care in the future, Potapenko and his colleagues stressed the importance of new approaches for more effective patient management. They presented a new design for an AMD patient tracking system based on a combination of temporally aware artificial intelligence (AI) and deterministic logic. The framework would autonomously suggest the treatment of the patient according to an observation and planning (O&P) regimen and address several advanced concepts.
The AI system was designed to make treatment decisions through the evaluation of clinical and imaging data from current and previous exams. It consisted of two main components: a deep learning-based model designed to detect CNV activity on optical coherence tomograms (OCT) and a deterministic logic layer that defines the impact of its output on patient management and treatment.
Investigators prospectively collected a data set of 200 real-world AMD follow-ups, including the treatment decision made by clinicians in our department. Each case was reassessed by retina specialists and the agreement between the AI decision and the expert consensus was compared to the initial treatment decision. From there, the investigators attempted to determine the proportion of patients that could be monitored autonomously by the proposed AI system.
The results indicate that unanimous agreement was reported in 46% of cases, higher in passively observed eyes (56%) than in actively treated eyes (36%; P = 0.007). The data show that the temporal AI model was superior in detecting disease activity compared to the model without temporal input (area under the curve [AUC]0.900 [95% CI, 0.894 – 0.906] and 0.857 [95% CI, 0.846 – 0.867]respectively).
The new models with and without fundus photography also had higher AUC sensitivity (AUC, 0.836 [95% CI, 0.827 – 0.845]) and 0.837 (95% CI, 0.828 – 0.846) versus 0.762 (95% CI, 0.746 – 0.778), respectively. Then, in validation against expert consensus, the data shows that the full AI system made safe autonomous decisions in 67% (n=134) and unsafe decisions in 6% (n=12) of eyes. , while a second opinion was requested in 27% of cases. (n = 54) cases.
Moreover, the AI system made the same decision as the expert consensus in 89.2% (n=157) of the cases, while the retina clinical agreement rate with expert consensus was 85, 8% (n=151; P = 0.42). AI-based tracking was able to make an autonomous decision in 73% of cases, of which 91.8% agreed with expert consensus.
Investigators noted that it was tied with the 87.7% concordance rate between decisions made at the clinic and expert consensus (P = 0.33).
“Even if not deployed to its full potential, the proposed algorithm could greatly alleviate the pressure placed on public ophthalmology services by an increasing number of AMD patients and be part of an effective system of follow-up and treatment in collaboration with the primary sector services,” concluded Potapenko.
The study, “Automated artificial intelligence-based system for the clinical monitoring of patients with age-related macular degeneration”, was published in Acta Ophthalmologica.
#Autonomous #AIbased #system #compliant #expert #consensus #treatment #AMD