Dr. Arun Krishnan - Virtual Computer Aide for Doctors
In a just short period of time, an imaging scan can produce hundreds of images, analyzing every section of the body, in search of suspicious structures. “The clinician has to be extremely attentive to subtle details in every study,” says Dr. Arun Krishnan, 43. However, there isn’t additional time during the day to analyze the flood of images. The physician first needs to search the images for all suspicious structures and then provide a careful interpretation to characterize them. “We need to keep in mind that findings in the images can help differentiate between benign and malignant lesions, and make a world of difference to the patient,“ says Krishnan. “Doctors want help. As I watched doctors in clinical practice managing their daily workload, I wondered how we could aid them.”
A native of India, Krishnan joined Siemens in 1995 as an employee of Siemens Corporate Research in Princeton, New Jersey, USA. In 2002, he was one of the co-founders of the CAD Group, an internal start-up within Siemens Medical Solutions USA, Inc., which was established to develop Computer-Aided Detection (CAD) products. Utilizing the rich information contained in medical images, CAD serves as a virtual computer aide who searches the images for suspicious patterns. The software is created by researchers well-versed in artificial intelligence techniques of image-processing and machine-learning. These researchers analyze large databases of images and their corresponding interpretation by expert radiologists.
Krishnan and his team of researchers and software engineers have worked with world renowned luminary clinical collaborators to create numerous CAD products that are used in hundreds of hospitals worldwide. For example, a CAD system for the early detection of lung lesions can detect nodules ranging from three to 30 millimeters in diameter. “When the doctor is finished with the analysis of the chest CT images, he or she can use this program to check if something was overlooked,” explains Krishnan. In addition to research and development, there is also a need for extensive clinical validation to demonstrate the additive value. Siemens has conducted one of the largest multi-reader, multi-case studies on the use of Lung CAD in chest CT. The study showed that all participating readers improved their diagnostic accuracy significantly with the assistance of CAD. Similar products developed by the CAD Group are used in digital chest X-ray, digital mammography and virtual colonography.
The next generation of CAD solutions, where Krishnan is also playing a major role, is taking the already innovative technology closer to the diagnosis realm, while becoming more comprehensive in the types of diseases that a CAD system will be able to detect. “To do this, we need to develop software that can answer the same questions the doctor asks,” says Krishnan when explaining how such programs are designed. Large clinical repositories are required to create disease specific sub-populations, and learn image features that correlate with the presence/absence of a disease. “We need to analyze massive amounts of data from hospitals plus the findings, and develop patterns for different diseases,” says Krishnan. Ideally, the system will provide the doctor with a series of possible diagnoses that could be relevant to the clinical case. “The system has to predict what the doctor needs,” adds Krishnan.
“Even in the future, CAD tools won’t be able to replace the diagnosis by a doctor,” cautions Krishnan. The difficulty when developing CAD solutions is striking the proper balance. “If it turns up suspicious findings with every normal study, the doctor will no longer take it seriously. On the other hand, the doctor must be confident that the system does not ignore important facts,” explains Krishnan. He and his team are working hard to strike the right balance and help the doctors provide the best possible care.
Krishnan’s colleagues include several researchers who have also won awards for their papers and inventions. Krishnan greatly appreciates the working atmosphere, which crackles with the spirit of invention, and he is proud of the award, for which he shares the credit with his world-class team. “Most of our inventions are collaborative efforts,” he says. Based on his 59 inventions 52 IPR families have been awarded. At the moment Krishnan holds 21 patents.
Disclaimer: On account of certain regional limitations of sales rights and service availability, Siemens cannot guarantee that the products included in this document are available through the Siemens sales organization worldwide. Availability and packaging may vary by country and are subject to change without prior notice. Some/All of the features and products described herein may not be available in the United States. The information in this document contains general technical descriptions of specifications and options as well as standard and optional features which do not always have to be present in individual cases. Siemens reserves the right to modify the design, packaging, specifications, and options described herein without prior notice. Please contact your local Siemens sales representative for the most current information.