Artificial Intelligence vs. Tuberculosis
Women in the Startup Environment Promise Little and Over Deliver, Says Qure.ai Founder Pooja Rao
We had a great RSNA 2018, with 4 scientific presentations, interesting conversations at our booth in the machine learning showcase, and demos at our partners’ booths.
Qure.ai won the AI Game changer award at the 6th NASSCOM Big Data & Analytics Summit on 11th July 2018. The award aimed to recognize & showcase the most innovative, high impact and high-tech AI solutions. Qure.ai’s qER solution is selected among the 273 case studies submitted by 163 companies.
Our chest x-ray product, qXR has received CE certification. Trained on more than one million chest x-rays, qXR detects 15 of the most common chest x-ray abnormalities with an accuracy of more than 90 percent. Taking only milliseconds to run, the product uses a heat map or bounding box to point out abnormalities to the clinician, facilitating rapid conﬁrmation. In settings without trained healthcare professionals, qXR is already being used to screen for tuberculosis, proving a valuable supplement to the existing healthcare systems. There are fewer than five companies globally who have received a CE certification for AI-based radiology products.
“The chest x-ray is the most commonly-performed radiology investigation, but one of the toughest to interpret,” said Dr. Shalini Govil, Quality Controller for the Columbia Asia Radiology Group. “Qure.ai’s solution could serve as a radiology assistant, providing a draft report that can be validated by a physician or radiologist. says. They’ve also come up with technology to visualize what the algorithm sees – a way to ‘see through the computer’s eyes.’ I think this will be a game-changer on the road to building confidence in AI.”
This technology is also being used globally as a screening tool for Tuberculosis. Forty percent of the 10.4 million annual Tuberculosis cases do not get proper care because they’re not properly diagnosed. qXR can substantially speed up the Tuberculosis diagnosis protocol by screening chest X-rays in milliseconds. More details about qXR and its use-cases are available at www.qure.ai/qxr.
“We’re excited to announce this certification, which clears our path to market in many geographies,” said Prashant Warier, Co-Founder and CEO of Qure.ai. “qXR can help doctors quickly and accurately detect and highlight abnormalities, reducing the chances of a missed diagnosis.”
We have also released a web-based interface that allows radiologists to test the performance of qXR. This portal – available at scan.qure.ai – gives users a real-time interpretation of a chest x-ray.
Qure.ai will be at SIIM 2018. Come by the start-up showcase (booth #113) to learn more about our products and research, or get a live demo
We have partnered with Teleradiology Solutions, a global pioneer in remote radiology interpretation and telehealth, and Telerad Tech (T2), a global health IT company and AI-enabled RIS-PACS provider, to enable smarter and faster diagnoses of X-ray and CT scan data, and reduce costs.
Through this partnership, Qure.ai’s chest X-ray technology will be integrated with Telerad Tech’s proprietary RIS PACS platform – RADSpa that TRS uses to provide teleradiology services globally.
“At Qure.ai, we are expanding the reach of our AI algorithms to help medical professionals deliver better outcomes to their patients,” said Prashant Warier, Co-Founder and CEO, Qure.ai. “TRS and Telerad Tech are pioneers in their respective domains, and we are excited about the impact this partnership will have on the millions of patients’ lives they touch.”
TRS has reported scans for over five million patients since inception in 2002, and currently caters to the requirements of 150 hospitals and healthcare centers in more than 20 countries, including the United States, Singapore, Nigeria, Tanzania, Uganda, Maldives and India.
Telerad Tech’s RADSpa platform is deployed in 24 countries and has processed more than 20 million studies and billions of images. RADSpa is FDA approved and CE certified.
TRS’s clinical expertise and Telerad Tech’s RADSpa platform consolidates radiology report information from diverse sources so radiologists can review these reports from anywhere and anytime. By integrating Qure.ai’s algorithms that automatically generate abnormality reports from X-rays and CT scans, radiologists and hospitals using RADspa will now have cutting-edge algorithms at their disposal to help prioritize care, make smarter and faster diagnoses, and reduce costs. This integration is expected to go live in the next four months in several Indian states where TRS provides teleradiology services.
“Making sure that doctors and hospitals have the necessary and highest quality information has to be the topmost priority,” said Dr Arjun Kalyanpur, CEO Teleradiology Solutions. “Diseases that are of public health importance worldwide such as tuberculosis, are within our focus of interest with a goal of providing high-quality diagnostics to facilitate early detection. Qure.ai, TRS and T2 have a shared vision when it comes to achieving this goal, and we strongly believe this will be highly beneficial for doctors, hospitals and patients alike.”
Qure.ai’s chest X-ray solution can automatically screen for abnormal chest X-rays and tuberculosis. Qure.ai’s automated reads can have a substantial impact on the screening protocol for tuberculosis globally.
“Qure.ai’s chest X-ray solution helps in early identification of probable TB cases and helps doctors in fast-tracking of TB patients for confirmatory diagnosis. It will act as a force multiplier for early and fast detection,” says Dr. Shibu Vijayan, Director at PATH India, an NGO that has been working to improve TB outcomes in India for years.
Today we launched a new AI-powered technology to accurately identify bleeds, fractures and other critical abnormalities in head CT scans. We have also released the results of an unprecedented clinical validation study confirming the algorithms’ near-radiologist performance on 21,000 patients and made a dataset of almost 500 AI-analyzed head CT scans available for download.
We trained the new AI using a collection of 313,318 anonymized head CT scans, along with their corresponding clinical reports. Of these, 21,095 scans were then used to validate the algorithms. Finally, the AI was clinically validated on 491 CT scans, with the results compared against a panel of three senior radiologists. The panel of radiologists included Norbert G. Campeau, M.D., a senior neuro-radiologist from the Mayo Clinic’s Department of Radiology. The validation study found that our AI was more than 95% accurate in identifying abnormalities.
In addition to the study, We have made a dataset of 491 AI-interpreted head CT scans, as well as the corresponding interpretations from the three radiologists, publicly available for download. This dataset is from the Centre for Advanced Research in Imaging, Neurosciences and Genomics, and includes both out-patient and in-patient scans from 7 centers. To download the full dataset, visit: http://headctstudy.qure.ai.
We have 6 scientific presentations at the ECR this year – in the ‘Deep Learning’, ‘Artificial intelligence in chest imaging’, and ‘Deep learning and Radiomics’ sessions. We will also be showcasing our new Chest X-ray and Head CT product demos.
We will be exhibiting at Booth #8564, Machine Learning showcase. To learn more about Qure’s research, you can also attend these scientific presentations:
- Generating Heatmaps to visualize the evidence of deep learning based diagnosis of Chest X-rays</strong>; Artificial Intelligence and Deep Learning in Medical Imaging, 11/26/17, 11:45 AM – 11:55 AM, Room S403A.
- Deep neural networks to identify and localize intracerebral hemorrhage and midline shift in brain CT scans</strong>; Informatics Sunday Poster Discussions, 11/26/17, 1:00 PM – 1:30 PM, INS-SUB | IN Community Learning Center.