EnvoyAI Partner:


Qure Green AI logo-01.png
if_Laptop_B_104985.png
Website: www.qure.ai/
if_290_279051.png
Year Founded: 2016
if_search_244366.png
Area of Focus/Speciality: Clinical Specialty - Radiology.
Body Regions - Chest, Head, Knee.
if_hospital_clinic_133105.png
Hospital Associations: N/A
if__location_1964213.png
Company Headquarters: Maharashtra, India
2538894-128.png
Number of Algorithms: 2 algorithms - Head CT scan solution and Chest X-ray solution.
Qure.ai linkedin page
Qure.ai twitter profile

Qure.ai

About Qure.ai

Qure.ai develops cutting-edge deep learning technology that helps detect and highlight abnormalities in medical images. Qure’s algorithms precisely quantify disease or tumor volumes so that patient response to therapy can be monitored closely. We develop technologies that can identify potential abnormalities in X-rays, CT Scans and MRIs. Currently, we focus on identifying and highlighting a) Hemorrhages, Fracture, Atrophy, Pneumocephalus, Infarcts, Contusions, Midline Shifts and Mass Effect in Head CT scans as well as prioritizing Radiologists' worklists, and b) identifying and highlighting more than 15 of the most common abnormalities in Chest Radiographs as well as assisting in triaging Chest X-rays.

Our mission is to make healthcare more affordable and accessible through the power of deep learning. 

if__location_1964213.png

Reference Sites:

Our solutions have been deployed at 4 different sites. We do have a few limitations in sharing this at the moment due to Non-Disclosure terms with existing partners.

Qure.ai Leadership
Dr. Prashant Warier, CEO

Dr. Prashant Warier

CEO

Dr. Pooja Rao, Co-Founder and Head of R&D

Dr. Pooja Rao

Co-Founder and Head of R&D

Qure.ai Machine | Qure.ai Head CT Assistant

Qure.ai is directly responsible for product and clinical representations

  • Example1.png
  • Example3.png
  • qure aI brain.png
  • qure ai brain2.png

Qure’s head CT product is designed for use in emergency care setting to help Radiologists prioritize patients and speed up decision-making. The solution has the capability to flag up critical scans for immediate review; automatically detects and localizes brain hemorrhages and fractures, and indicates the degree of severity of brain injury. It detects and quantifies the 4 types of intracranial hemorrhages (epidural, subdural, subarachnoid and intra-parenchymal), as well as cranial fractures and contusions. The algorithm also identifies the degree of severity and indicates if there is a mass effect or midline shift consequent to the injury. Each abnormality is quantified and the anatomical location is identified – generating a fully automated prefilled, editable report.

  • Modality: CT
  • Body Part Examined: Head
  • Study Description: Axial plane images of the from skull base to the vertex
  • Slice thickness: 5 mm
  • Applicable Study Types: The HEAD CT scan solution is designed to identify, highlight Bleeds and Fractures. The software application automatically detects Hemorrhages in the brain regions, quantify the hemorrhage volume and midline shift and mass-effect. 

Qure solutions have been trained on >1Million data from multiple sources. The data includes images captured from all leading machine manufacturers. Currently, we have performance of AUC more than 0.85 for all our solutions. The models used are convolutional neural networks.

Indications for use: The solutions are designed as an aid to focus the attention of the Radiologist by identifying, and marking abnormalities to improve efficiency, accuracy and turnaround time. Such conclusions are used for reporting the findings and passed on to physicians to carry out further treatment plan.

Qure does not require Personal Identifiable Information (PII) or patient data to perform it’s processing.

It is preferred that Clients send DICOMs with anonymized IDs.

Qure expects a DICOM push from the Client’s end.

The solution recommends the interpretation of results by a board-certified radiologist only.

Qure.ai Machine | Qure.ai CXR Assistant

Qure.ai is directly responsible for product and clinical representations

  • cardioM2.jpeg
  • CXR.jpeg
  • qure aI CXR.png

The Chest X-ray software solution dynamically details and highlights abnormal areas once identified. The solution highlights specific areas with abnormalities using bounding boxes. The solution capability expands to identifying frequently and rarely occurring abnormalities to screening of significant disease like Tuberculosis.


List of tags identified by Qure Chest X-ray solution are:

  • Atelectasis
  • Cervical Rib
  • Prominent hilum
  • Pneumothorax
  • Blunted CP
  • Consolidation
  • Pulmonary Edema
  • Calcification
  • Emphysema
  • Opacity
  • Reticulo Nodular Pattern
  • Cardiomegaly
  • Fibrosis
  • Tracheal Shift
  • Cavity
  • Pleural Effusion
  • Normal

Qure solutions have been trained on >1Million data from multiple sources. The data includes images captured from all leading machine manufacturers. Currently, we have performance of AUC more than 0.85 for all our solutions. The models used are convolutional neural networks.

Indications for use: The solutions are designed as an aid to focus the attention of the Radiologist by identifying, and marking abnormalities to improve efficiency, accuracy and turnaround time. Such conclusions are used for reporting the findings and passed on to physicians to carry out further treatment plan.

Qure does not require Personal Identifiable Information (PII) or patient data to perform it’s processing.

It is preferred that Clients send DICOMs with anonymized IDs.

Qure expects a DICOM push from the Client’s end.

The solution recommends the interpretation of results by a board-certified radiologist only.