Artificial intelligence (AI) aims to mimic human cognitive functions. It is bringing a paradigm shift to healthcare, powered by increasing availability of healthcare data and rapid progress of analytics techniques. We survey the current status of AI applications in healthcare and discuss its future. AI can be applied to various types of healthcare data (structured and unstructured). Popular AI techniques include machine learning methods for structured data, such as the classical support vector machine and neural network, and the modern deep learning, as well as natural language processing for unstructured data. Major disease areas that use AI tools include cancer, neurology and cardiology. We then review in more details the AI applications in stroke, in the three major areas of early detection and diagnosis, treatment, as well as outcome prediction and prognosis evaluation. We conclude with discussion about pioneer AI systems, such as IBM Watson, and hurdles for real-life deployment of AI.

The most effective medical treatment for acute ischaemic stroke (AIS) is to offer intravenous thrombolysis during the ultra-early period of time after the onset. Even based on the Consensus of Chinese Experts on Intravenous Thrombolysis for AIS in 2012 and 2014 Chinese Guidelines on the Diagnosis and Treatment of AIS, the rate of thrombolysis for AIS in China remained around 2.4%, and the rate of intravenous tissue plasminogen activator usage was only about 1.6% in real world. The indication of thrombolysis for AIS has been expanded, and contraindications have been reduced with recently published studies. In order to facilitate the standardisation of treating AIS, improve the rate of thrombolysis and benefit patients who had a stroke, Chinese Stroke Association has organised and developed this scientific statement.

Exogenous stem cell therapy (SCT) has been recognized recently as a promising neuroregenerative strategy to augment recovery in stroke survivors. Mesenchymal stem cells (MSCs) are the primary source of stem cells used in the majority of both pre-clinical and clinical studies in stroke. In the absence of evidence-based guidelines on the use of SCT in stroke patients, understanding the progress of MSC research across published studies will assist researchers and clinicians in better achieving success in translating research. We conducted a systematic review on published literature using MSCs in both pre-clinical studies and clinical trials between 2008 and 2017 using the public databases PubMed and Ovid Medline, and the clinical trial registry ( www.clinicaltrials.gov ). A total of 78 pre-clinical studies and eight clinical studies were identified. While majority of the pre-clinical and clinical studies demonstrated statistically significant effects, the clinical significance of these findings was still unclear. Effect sizes could not be measured mainly due to reporting issues in pre-clinical studies, thus limiting our ability to compare results across studies quantitatively. The overall quality of both pre-clinical and clinical studies was sub-optimal. By conducting a systematic review of both pre-clinical and clinical studies on MSCs therapy in stroke, we assessed the quality of current evidence and identified several issues and gaps in translating animal studies to human trials. Addressing these issues and incorporating changes into future animal studies and human trials may lead to better success of stem cells-based therapeutics in the near future.

Aim Spontaneous subarachnoid haemorrhage (SAH) caused by ruptured cerebral aneurysm is a severe subtype of haemorrhagic stroke. Although the incidence of SAH is relatively low among all cerebrovascular diseases, the mortality is the highest. The critical management of SAH is challenging. We provide this evidence-based guideline to present current and comprehensive recommendations for the diagnosis and treatment of non-trauma SAH. Methods A formal literature search of MEDLINE (1 January 1990–30 June 2019) was performed. Data were synthesised with the use of evidence tables. Writing group members met by teleconference to discuss data-derived recommendations. The Chinese Stroke Association’s levels of evidence grading algorithm was used to grade each recommendation. The guideline draft was reviewed by Chinese Stroke Association’s Stroke Fellow Committees. It is intended that this guideline be fully updated every 3 years. Results Evidence-based guidelines are presented for the care of patients presenting with non-trauma SAH. The focus of the guideline was subdivided into transfer and systems of care, diagnosis flowchart, aetiology and differentiation, prevention of rebleeding, surgical and endovascular repair of ruptured aneurysms, management of vasospasm and delayed cerebral ischaemia, management of hydrocephalus, management of seizures and management of medical complications. Conclusions The guideline offers a framework for SAH management. Early professional and aggressive care of SAH might help dramatically.