Ultrasound and other medical imaging devices generate images by scanning biological structures and/or tissues of patients. These images aid in the diagnosis of serious illnesses such as cancer. Issues such poor quality and clarity of scanned images or conditions associated with the patient during scans or skill of the technician can lead to incorrect or misdiagnosis.
Dr. Rick Mammone and Dr. Christine Podilchuk, researchers at Rutgers University, have developed a novel method to assist in the identification of objects of interest in medical images. The method utilizes deep learning neural networks to automatically identify and highlight suspicious objects in the images drawing a bounding box around regions of interest. The technology provides diagnosticians the ability to use a single-image method to find the regions of interest in real-time, enabling quick identification of suspicious lesions exhibiting the highest probability of malignancy or other disease states.
Images obtained directly from an imaging device such as an ultrasound machine, a mobile device, or images that have already been captured and stored are processed by the technology.
This technology can be used for any type of medical image display and diagnostic technology (software, device, other.)
In addition, the technology can be bundled with four related technologies to create a complete, cost-effective, and accurate solution for enhancing, transmitting, and interpreting and providing diagnoses using captured medical images. The other technologies include:
- Easy identification of areas of concern
- Device and software independent
- For diagnosis of a wide array of medical conditions using medical images
- Supports images generated by, stored, or captured by imaging devices and a unique ability to capture and enhance medical images on a mobile phone
- Telemedicine / Teleradiology
- Mobile medical scanning (e.g., mobile ultrasound)
- Medical Image Solutions, Systems & Software
- Electronic Medical/Health Systems (EMR/EHR)
Intellectual Property & Development Status:
US Patent 11,043,297; Available for use with new or existing image display/diagnostic applications or within a complete system comprised of additional related technologies from the inventors. We are seeking licensing and/or industry partners.