Our engine of research and innovation is powered by a large multidisciplinary team comprising of deep learning talents, top ranking hospitals and advisors from leading universities. With the same vision and collaborative spirit, we jostle with some of the world’s most complex healthcare topics to achieve breakthrough technologies to pave the way for the future of healthcare. The ultimate beneficiaries include patients, clinicians and healthcare organisations on a global scale.
small Vessel Diseases
Brain Age Estimation
Artificial Intelligence Can Effectively Predict Early Hematoma Expansion of Intracerebral Hemorrhage Analysing Non-contrast Computed Tomography Image
Compared with the existing prediction methods through computed tomographic angiography (CTA) image features and NCCT image features analysis, our AI model has higher specificity and sensitivity in the prediction of early hematoma enlargement in patients with ICH. This provides a time-saving, easy to implement and subjective independent method to predict the risk of hematoma enlargement.
Primary Categorizing and Masking Cerebral Small Vessel Disease Based on “Deep Learning System”
The results of our comparison lend support to the case that an appropriately trained DLS can be trusted to the same extent as one would trust an attending doctor with a few years of experience for both segmentation and detection of lesions. According to the current results, the model can obtain lesion recognition and segmentation at the level of attending physicians. The analysis can be performed over a hundred times faster than the physicians, thereby reducing the repetitive labor of physicians. Besides the categorization and masking of the CSVD, the software determines the volume of the lesion, which is critical to making decisions for dosage prescription, among other clinical decisions.
Brain Age Estimation from MRI using Cascade Networks with Ranking Loss
The proposed ensemble TSAN after bias correction generally yields the best classification performance in terms of AUC, accuracy, sensitivity and specificity. It confirms that brain age (i.e. the estimated brain age) is a promising biomarker for dementia classification or early stage dementia risk screening. With this tool doctors are able to identify patients with a higher risk of age-related brain deteoration and decide which treatment is the most effective to fight such risk before the onset of cognitive decline.
Deep Neural Network-based Detection and Segmentation of Intracranial Aneurysms on 3D Rotational DSA
A combination of 3D-Dense-UNet model and 3D-RA images could achieve high sensitivity in IA detection with a relatively low false positive rate, leading to faster diagnostics. Meanwhile, maximal diameter derived from prediction mask showed good clinical application prospect, as it is a consistent method to measure IA, which is comparable between patients and does not include observer bias. Overall, this system offers an improved patient risk-stratification, enabling the doctors to select the optimal management of IA.
Assisting Scalable Diagnosis Automatically via CT images in the Combat Against COVID‑19
COVIDNet offered one powerful tool for screening the COVID-19 suspected patients. As of March 23, 2020, the COVIDNet system had been employed in 6 hospitals in China with PCR confirmation. COVIDNet achieved an accuracy rate of 94.3%, as compared to the average accuracy rate of 83.9% of COVID-19 trained radiologists and average accuracy rate of 78.9% of non-COVID-19 trained radiologists. COVIDNet can increase the diagnostic accuracy and reduce time to diagnosis by providing addictional information for radiologists, allowing a better management of patients