-
Processing of Electronic Health Records using Deep Learning: A review
Authors:
Venet Osmani,
Li Li,
Matteo Danieletto,
Benjamin Glicksberg,
Joel Dudley,
Oscar Mayora
Abstract:
Availability of large amount of clinical data is opening up new research avenues in a number of fields. An exciting field in this respect is healthcare, where secondary use of healthcare data is beginning to revolutionize healthcare. Except for availability of Big Data, both medical data from healthcare institutions (such as EMR data) and data generated from health and wellbeing devices (such as p…
▽ More
Availability of large amount of clinical data is opening up new research avenues in a number of fields. An exciting field in this respect is healthcare, where secondary use of healthcare data is beginning to revolutionize healthcare. Except for availability of Big Data, both medical data from healthcare institutions (such as EMR data) and data generated from health and wellbeing devices (such as personal trackers), a significant contribution to this trend is also being made by recent advances on machine learning, specifically deep learning algorithms.
△ Less
Submitted 5 April, 2018;
originally announced April 2018.
-
Smartphone apps usage patterns as a predictor of perceived stress levels at workplace
Authors:
Raihana Ferdous,
Venet Osmani,
Oscar Mayora
Abstract:
Explosion of number of smartphone apps and their diversity has created a fertile ground to study behaviour of smartphone users. Patterns of app usage, specifically types of apps and their duration are influenced by the state of the user and this information can be correlated with the self-reported state of the users. The work in this paper is along the line of understanding patterns of app usage a…
▽ More
Explosion of number of smartphone apps and their diversity has created a fertile ground to study behaviour of smartphone users. Patterns of app usage, specifically types of apps and their duration are influenced by the state of the user and this information can be correlated with the self-reported state of the users. The work in this paper is along the line of understanding patterns of app usage and investigating relationship of these patterns with the perceived stress level within the workplace context. Our results show that using a subject-centric behaviour model we can predict stress levels based on smartphone app usage. The results we have achieved, of average accuracy of 75% and precision of 85.7%, can be used as an indicator of overall stress levels in work environments and in turn inform stress reduction organisational policies, especially when considering interrelation between stress and productivity of workers.
△ Less
Submitted 10 March, 2018;
originally announced March 2018.
-
Enabling Prescription-based Health Apps
Authors:
Venet Osmani,
Stefano Forti,
Oscar Mayora,
Diego Conforti
Abstract:
We describe an innovative framework for prescription of personalised health apps by integrating Personal Health Records (PHR) with disease-specific mobile applications for managing medical conditions and the communication with clinical professionals. The prescribed apps record multiple variables including medical history enriched with innovative features such as integration with medical monitoring…
▽ More
We describe an innovative framework for prescription of personalised health apps by integrating Personal Health Records (PHR) with disease-specific mobile applications for managing medical conditions and the communication with clinical professionals. The prescribed apps record multiple variables including medical history enriched with innovative features such as integration with medical monitoring devices and wellbeing trackers to provide patients and clinicians with a personalised support on disease management. Our framework is based on an existing PHR ecosystem called TreC, uniquely positioned between healthcare provider and the patients, which is being used by over 70.000 patients in Trentino region in Northern Italy. We also describe three important aspects of health app prescription and how medical information is automatically encoded through the TreC framework and is prescribed as a personalised app, ready to be installed in the patients' smartphone.
△ Less
Submitted 29 June, 2017;
originally announced June 2017.
-
Automatic Stress Detection in Working Environments from Smartphones' Accelerometer Data: A First Step
Authors:
Enrique Garcia-Ceja,
Venet Osmani,
Oscar Mayora
Abstract:
Increase in workload across many organisations and consequent increase in occupational stress is negatively affecting the health of the workforce. Measuring stress and other human psychological dynamics is difficult due to subjective nature of self- reporting and variability between and within individuals. With the advent of smartphones it is now possible to monitor diverse aspects of human behavi…
▽ More
Increase in workload across many organisations and consequent increase in occupational stress is negatively affecting the health of the workforce. Measuring stress and other human psychological dynamics is difficult due to subjective nature of self- reporting and variability between and within individuals. With the advent of smartphones it is now possible to monitor diverse aspects of human behaviour, including objectively measured behaviour related to psychological state and consequently stress. We have used data from the smartphone's built-in accelerometer to detect behaviour that correlates with subjects stress levels. Accelerometer sensor was chosen because it raises fewer privacy concerns (in comparison to location, video or audio recording, for example) and because its low power consumption makes it suitable to be embedded in smaller wearable devices, such as fitness trackers. 30 subjects from two different organizations were provided with smartphones. The study lasted for 8 weeks and was conducted in real working environments, with no constraints whatsoever placed upon smartphone usage. The subjects reported their perceived stress levels three times during their working hours. Using combination of statistical models to classify self reported stress levels, we achieved a maximum overall accuracy of 71% for user-specific models and an accuracy of 60% for the use of similar-users models, relying solely on data from a single accelerometer.
△ Less
Submitted 14 October, 2015;
originally announced October 2015.
-
Smartphones in Mental Health: Detecting Depressive and Manic Episodes
Authors:
Venet Osmani,
Agnes Gruenerbl,
Gernot Bahle,
Christian Haring,
Paul Lukowicz,
Oscar Mayora
Abstract:
An observational study with patients diagnosed with bipolar disorder investigates whether data from smartphone sensors can be used to recognize bipolar disorder episodes and detect behavior changes that can signal an onset of an episode using objective data.
An observational study with patients diagnosed with bipolar disorder investigates whether data from smartphone sensors can be used to recognize bipolar disorder episodes and detect behavior changes that can signal an onset of an episode using objective data.
△ Less
Submitted 19 October, 2015; v1 submitted 6 October, 2015;
originally announced October 2015.
-
Investigation of indoor localization with ambient FM radio stations
Authors:
Andrei Popleteev,
Venet Osmani,
Oscar Mayora
Abstract:
Localization plays an essential role in many ubiquitous computing applications. While the outdoor location-aware services based on GPS are becoming increasingly popular, their proliferation to indoor environments is limited due to the lack of widely available indoor localization systems. The de-facto standard for indoor positioning is based on Wi-Fi and while other localization alternatives exist,…
▽ More
Localization plays an essential role in many ubiquitous computing applications. While the outdoor location-aware services based on GPS are becoming increasingly popular, their proliferation to indoor environments is limited due to the lack of widely available indoor localization systems. The de-facto standard for indoor positioning is based on Wi-Fi and while other localization alternatives exist, they either require expensive hardware or provide a low accuracy. This paper presents an investigation into localization system that leverages signals of broadcasting FM radio stations. The FM stations provide a worldwide coverage, while FM tuners are readily available in many mobile devices. The experimental results show that FM radio can be used for indoor localization, while providing longer battery life than Wi-Fi, making FM an alternative to consider for positioning.
△ Less
Submitted 6 August, 2013;
originally announced August 2013.
