Artificial intelligence (AI) has been positioned as being the most important recent advancement in radiology, if not the most potentially disruptive. Singapore radiologists have been quick to embrace this technology as part of the natural progression of the discipline toward a vision of how clinical medicine, empowered by technology, can achieve our national healthcare objectives of delivering value-based and patient-centric care. In this article, we consider 3 core questions relating to AI in radiology, and review the barriers to the widespread adoption of AI in radiology. We propose solutions and describe a "Centaur" model as a promising avenue for enabling the interfacing between AI and radiologists. Finally, we introduce The Radiological AI, Data Science and Imaging Informatics (RADII) subsection of the Singapore Radiological Society. RADII is an enabling body, which together with key technological and institutional stakeholders, will champion research, development and evaluation of AI for radiology applications.
Artificial Intelligence ; Humans ; Image Processing, Computer-Assisted ; Machine Learning ; Neural Networks (Computer) ; Radiology ; Singapore ; Societies, Medical

INTRODUCTIONIn end-stage heart failure (HF) that is not eligible for mechanical assist device or heart transplant, palliative care serves to maximise symptom control and quality of life. We sought to evaluate the impact of home-based advance care programme (ACP) on healthcare utilisation in end-stage HF patients.

MATERIALS AND METHODSProspectively collected registry data on all end-stage HF recruited into ACP between July 2008 and July 2010 were analysed. Chart reviews were conducted on HF database and hospital electronic records. Phone interview and home visit details by ACP team were extracted to complete data entry. HF and all-cause hospitalisations 1 year before, and any time after ACP inception were defined as events. For the latter analysis, follow-up duration adjustment to event episodes was performed to account for death less than a year.

RESULTSForty-four patients (mean age 79 years, 39% men) were followed up for 15±8 months. Fifty-seven percent had diabetes, 80% ischaemic heart disease, and 60% chronic kidney disease. All reported functional class III/IV at enrolment. Mean serum sodium was 136±6 mmol/L, and creatinine 186±126 mmol/L. Thirty (68%) died within the programme. Mean time to death was 5.5 months. Mean all-cause and HF hospitalisations were 3.6 and 2.0 per patient before enrolment, but improved to 1.0 and 0.6 respectively after ACP. Thirty-six (71%) patients had fewer HF hospitalisations. When only those who survived more than a year were considered (n = 14), 10 (71%) and 9 (64%) experienced reduced HF (mean: 1.4 episodes per patient) and all-cause hospitalisations (mean: 2.2 episodes per patient) respectively.

CONCLUSIONHome-based advance care programme is potentially effective in reducing healthcare utilisation of end-stage HF patients, primarily by reducing HF rehospitalisations, and in probably saving costs as well.

Aged ; Aged, 80 and over ; Diabetes Mellitus ; Female ; Health Care Costs ; Health Services ; economics ; utilization ; Heart Failure ; complications ; economics ; therapy ; Home Care Services, Hospital-Based ; economics ; Hospitalization ; economics ; statistics & numerical data ; Humans ; Male ; Myocardial Ischemia ; complications ; Palliative Care ; economics ; methods ; Prospective Studies ; Registries ; Renal Insufficiency, Chronic ; complications ; Tertiary Care Centers