Category: Projects

Deep Representation Learning from Electronic Health Records

As Electronic Health Records (EHRs) become more ubiquitous, healthcare providers are beginning to appreciate the benefits of using this data for ‘secondary applications’, such as disease diagnosis and readmission prediction, by applying machine learning techniques (in particular, deep neural networks). The difficulty with this approach is that data from EHRs is often inconsistent, episodic, exists …

A Patient’s Journey for an Elective Surgery

An Automated Approach Towards Prioritisation and Preoperative Assessment The path to an elective surgery can contain many difficulties for both patients and clinicians. It is a lengthy, as well as daunting, process for the patient, while delays and cancellations can be extremely costly for the healthcare provider. This research project looked at how to devise …

App Analytics and the Questions They Answer

Summer of Research project by Tuan Dung Ngo, AUT, supervised by Associate Professor Robyn Whittaker (University of Auckland and Waitemata DHB). Mobile technology has developed rapidly over the last few years. Almost everyone you know has a smartphone. You might even be reading this article on a smartphone. As various smartphone applications rise and fall in …

Building a surgical outcome calculator for New Zealand patients

For patients and clinicians, deciding whether to go ahead with major surgery or not can be difficult. Your surgeon and anaesthetist will discuss both the benefits and risks of having surgery to help with the shared decision making process. In the past, many clinicians have found it difficult to accurately assess the risk of having …

information consent best practice

Information Consent Best Practice

Data-driven healthcare will result in a “tsunami” of information from existing and new data sources including patient-generated data from genetic testing, consumer devices such as wearable fitness apps, and social media. How can clinicians and researchers utilise this data to help consumers achieve better health outcomes? How can consumers control who has access to their …

HOPE: Health Outcomes Prediction Engine

Currently, treatment decisions for elective surgery are currently based on clinicians’ assessments of the benefits the surgery will bring to the patient. This assessment is derived from the clinicians’ personal experience and knowledge of published evidence of effectiveness. In New Zealand, eligibility for surgery is based on a prioritisation score derived from this clinical assessment …

Vital Sign Monitoring and Decision Support System

During admission to hospital, patients often show signs of acute physical deterioration before a serious event occurs. Current variation in vital signs charts, early warning scores, skills and knowledge of responders and availability of responders in hospitals means these patients may not receive timely, expert care when they need it, potentially leading to severe consequences. …

Making Existing Data Sources Available

New Zealand has excellent health data due to national ID number and high quality data scientists. We seek to link them together by making health data easier to access for research. We have reviewed publicly available health datasets in New Zealand and: (1) Produced a website to collate links to this data (2) Analysed relationships between …