No abstract

Background no studies have compared the predictive validity of different dementia risk prediction models in Australia. Objectives (i) to investigate the predictive validity of the Australian National University-Alzheimer’s Disease Risk Index (ANU-ADRI), LIfestyle for BRAin Health (LIBRA) Index and cardiovascular risk factors, ageing and dementia study (CAIDE) models for predicting probable dementia/cognitive impairment in an Australian cohort. (ii) To develop and assess the predictive validity of a new hybrid model combining variables from the three models. Methods the Hunter Community Study (HCS) included 3,306 adults aged 55–85 years with a median follow-up of 7.1 years. Probable dementia/cognitive impairment was defined using Admitted Patient Data Collection, dispensing of cholinesterase inhibitors or memantine, or a cognitive test. Model validity was assessed by calibration and discrimination. A hybrid model was developed using deep neural network analysis, a machine learning method. Results 120 (3.6%) participants developed probable dementia/cognitive impairment. Mean calibration by ANU-ADRI, LIBRA, CAIDE and the hybrid model was 19, 0.5, 4.7 and 3.4%, respectively. The discrimination of the models was 0.65 (95% CI 0.60–0.70), 0.65 (95% CI 0.60–0.71), 0.54 (95% CI 0.49–0.58) and 0.80 (95% CI 0.78–0.83), respectively. Conclusion ANU-ADRI and LIBRA were better dementia prediction tools than CAIDE for identification of high-risk individuals in this cohort. ANU-ADRI overestimated and LIBRA underestimated the risk. The new hybrid model had a higher predictive performance than the other models but it needs to be validated independently in longitudinal studies.

No abstract

This paper presents an end-to-end assistive EyeWear prototype aimed at Vision Impaired users. The prototype uses computer vision to detect objects on planar surfaces and sonifies their 3D locations using spatial audio. The prototype system is novel in that the wearable component and real-time operation of the system allows the user to interactively affect the audio feedback by actively and intuitively moving a headworn sensor. User trials were conducted on 12 blindfolded subjects who were tasked to perform an object localisation and placement task using our system. Quantitative trial results and qualitative user feedback suggest that the prototype has potential as a real world assistive device.

We present an end-to-end prototype for an assistive EyeWear system aimed at Vision Impaired users. The system uses computer vision to detect objects on planar surfaces and sonifies their 3D locations using spatial audio. A key novelty of the system is that it operates in real time (15Hz), allowing the user to interactively affect the audio feedback by actively moving a headworn sensor. A quantitative user study was conducted on 12 blindfolded subjects performing an object localisation and placement task using our system. This detailed study of near field interactive spatial audio for users operating at around arm's length departs from existing studies focused on far-field audio and non-interactive systems. The object localisation accuracy achieved on naive users suggests that the EyeWear prototype has a lot of potential as a real world assistive device. User feedback collected from exit surveys and mathematical modelling of user errors provide several promising avenues to further improve system performance.