Browsing by Author "Moilanen, Jukka"

Now showing 1 - 2 of 2

A comprehensive model for measuring real-life cost-effectiveness in eyecare: automation in care and evaluation of system (aces-rwm™)
(2022-05) Tuulonen, Anja; Kataja, Marko; Aaltonen, Vesa; Kinnunen, Kati; Moilanen, Jukka; Saarela, Ville; Linna, Miika; Malmivaara, Antti; Uusitalo-Jarvinen, Hannele
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
This paper describes a holistic, yet simple and comprehensible, ecosystem model to deal with multiple and complex challenges in eyecare. It aims at producing the best possible wellbeing and eyesight with the available resources. When targeting to improve the real-world cost-effectiveness, what gets done in everyday practice needs be measured routinely, efficiently and unselectively. Collection of all real-world data of all patients will enable evaluation and comparison of eyecare systems and departments between themselves nationally and internationally. The concept advocates a strategy to optimize real-life effectiveness, sustainability and outcomes of the service delivery in ophthalmology. The model consists of three components: (1) resource-governing principles (i.e., to deal with increasing demand and limited resources), (2) real-world monitoring (i.e., to collect structured real-world data utilizing automation and visualization of clinical parameters, health-related quality of life and costs), and (3) digital innovation strategy (i.e., to evaluate and benchmark real-world outcomes and cost-effectiveness). The core value and strength of the model lies in the consensus and collaboration of all Finnish university eye clinics to collect and evaluate the uniformly structured real-world outcomes data. In addition to ophthalmology, the approach is adaptable to any medical discipline to efficiently generate real-world insights and resilience in health systems.
DR-GPT : A large language model for medical report analysis of diabetic retinopathy patients
(2024-10) Jaskari, Joel; Sahlsten, Jaakko; Summanen, Paula; Moilanen, Jukka; Lehtola, Erika; Aho, Marjo; Säpyskä, Elina; Hietala, Kustaa; Kaski, Kimmo
A1 Alkuperäisartikkeli tieteellisessä aikakauslehdessä
Diabetic retinopathy (DR) is a sight-threatening condition caused by diabetes. Screening programmes for DR include eye examinations, where the patient’s fundi are photographed, and the findings, including DR severity, are recorded in the medical report. However, statistical analyses based on DR severity require structured labels that calls for laborious manual annotation process if the report format is unstructured. In this work, we propose a large language model DR-GPT for classification of the DR severity from unstructured medical reports. On a clinical set of medical reports, DR-GPT reaches 0.975 quadratic weighted Cohen’s kappa using truncated Early Treatment Diabetic Retinopathy Study scale. When DR-GPT annotations for unlabeled data are paired with corresponding fundus images, the additional data improves image classifier performance with statistical significance. Our analysis shows that large language models can be applied for unstructured medical report databases to classify diabetic retinopathy with a variety of applications.