Industry News

Researchers Use Machine Learning to Increase Resolution of OCT Imaging

By Medimaging International staff writers
03 Oct 2019

Image: A new technique, called optical coherence refraction tomography (OCRT), could improve medical images obtained in the multibillion-dollar OCT industry for medical fields ranging from cardiology to oncology (Photo courtesy of Pixabay).Biomedical engineers at Duke University (Durham, NC, USA) have devised a method for increasing the resolution of optical coherence tomography (OCT) down to a single micrometer scale in all directions. The new technique, called optical coherence refraction tomography (OCRT), could improve medical images obtained in the multibillion-dollar OCT industry for medical fields ranging from cardiology to oncology.

OCT is an imaging technology analogous to ultrasound that uses light rather than sound waves. A probe shoots a beam of light into a tissue and, based on the delays of the light waves as they bounce back, determines the boundaries of the features within. To get a full picture of these structures, the process is repeated at many horizontal positions over the surface of the tissue being scanned.

Since OCT provides much better resolution of depth than lateral direction, it works best when these features contain mostly flat layers. When objects within the tissue have irregular shapes, the features become blurred and the light refracts in different directions, reducing the image quality. Previous attempts at creating OCT images with high lateral resolution have relied on holography—painstakingly measuring the complex electromagnetic field reflected back from the object. While this has been demonstrated, the approach requires the sample and imaging apparatus to remain perfectly still down to the nanometer scale during the entire measurement.

However, the biomedical engineers at Duke University have taken a different approach. Instead of relying on holography, the researchers combined OCT images acquired from multiple angles to extend the depth resolution to the lateral dimension. Each individual OCT image, however, becomes distorted by the light’s refraction through irregularities in the cells and other tissue components. To compensate for these altered paths when compiling the final images, the researchers needed to accurately model how the light is bent as it passes through the sample.

To accomplish this computational feat, the biomedical engineers developed a method using “gradient-based optimization” to infer the refractive index within the different areas of tissue based on the multi-angle images. This approach determines the direction in which the given property—in this case, the refractive index—needs to be adjusted to create a better image. After several iterations, the algorithm creates a map of the tissue’s refractive index that best compensates for the light’s distortions. The method was implemented using TensorFlow, a popular software library created by Google for deep learning applications.

For proof-of-concept experiments, the researchers took tissue samples such as the bladder or trachea of a mouse, placed them in a tube, and rotated the samples 360 degrees beneath an OCT scanner. The algorithm successfully created a map of each sample’s refractive index, increasing the lateral resolution of the scan by more than 300% while reducing the background noise in the final image. While the study used samples already removed from the body, the researchers believe OCRT can be adapted to work in a living organism.

“One of the many reasons why I find this work exciting is that we were able to borrow tools from the machine learning community and apply them not only to post-process OCT images, but also to combine them in a novel way and extract new information,” said researcher Kevin Zhou. “I think there are many applications of these deep learning libraries such as TensorFlow and PyTorch, outside of the standard tasks such as image classification and segmentation.”

Related Links:
Duke University



E-mail Print
FaceBook Twitter Google+ Linked in

Additional news

25 Nov 2019
LANDAUER and RaySafe Exhibit New Products and Services at RSNA 2019
 LANDAUER (Glenwood, IL, USA) and RaySafe (Billdal, Sweden) showcased several new products and services at the Radiological Society of North America (RSNA) 2019 Annual Meeting held on December 1-5 at McCormick Place in Chicago, USA.
Read More
07 Nov 2019
Global Automated Breast Ultrasound Systems Market to Surpass USD 1.8 Billion by 2025
The global automated breast ultrasound systems (ABUS) market was valued at over USD 0.68 billion in 2018 and is projected to grow at a CAGR 16.1% during the forecast period, 2019-2025 to surpass USD 1.8 billion by 2025, driven by increasing government expenditure and rising awareness about breast cancer. Initiatives such as free check-up camps and symposiums for raising awareness regarding early detection and diagnosis of breast cancer will lead to greater adoption of ABUS in the coming years. However, stringent regulations and guidelines for the approval of ABUS, coupled with the high cost of diagnostic procedure, could hamper the market growth during the forecast period.
Read More
07 Nov 2019
CT System Receives FDA Clearance for AI-Based Image Reconstruction Technology
Canon Medical Systems USA, Inc. (Tustin, CA, USA; www.us.medical.canon) has received 510(k) clearance for its Advanced Intelligent Clear-IQ Engine (AiCE) for the Aquilion Precision, further expanding access to its new deep convolutional neural network (DCNN) image reconstruction technology. The technology, which is now available on Canon’s Aquilion Precision and Aquilion ONE/GENESIS Edition CT systems, uses a deep learning algorithm to differentiate signal from noise so that it can suppress noise while enhancing signal, thus improving CT image reconstruction.
Read More
07 Nov 2019
AI-Powered MRI Image Enhancement Software Receives FDA Clearance
A new image processing software that utilizes proprietary deep learning algorithms to bring the latest imaging enhancement technology to existing scanners has received 510(k) clearance from the US Food and Drug Administration (FDA). The software named SubtleMR uses denoising and resolution enhancement to improve image quality and has been developed by Subtle Medical (Menlo Park, CA, USA; www.subtlemedical.com), a healthcare technology company focused on improving the quality, efficiency, and accessibility of medical imaging.
Read More
06 Nov 2019
UCSF Launches Artificial Intelligence Center to Advance Medical Imaging
The University of California San Francisco {(UCSF) San Francisco, CA, USA; www.ucsf.edu} has announced the launch of a new center to accelerate the application of artificial intelligence (AI) technology to radiology, leveraging advanced computational techniques and industry collaborations to improve patient diagnoses and care. The Center for Intelligent Imaging, or ci2, will develop and apply AI to devise powerful new ways to look inside the body and to evaluate health and disease. The center aims to enable transformation via intelligent radiology, with the goal of again collaborating with the industry to become of the first institutions to bring medical imaging AI to the bedside.
Read More
06 Nov 2019
Global Blood Gas and Electrolyte Analyzers Market to Surpass USD 2.1 Billion by 2025
The global blood gas and electrolyte analyzers market is projected to grow at a CAGR of 6.5% during the forecast period, 2018-2025 to reach a value of over USD 2.1 billion by 2025, driven mainly by the Bench Top segment which is anticipated to bring in healthy gains and add significant momentum to the market growth. The Bench Top segment is projected to grow at a CAGR of over 6.4% during the forecast period.
Read More
06 Nov 2019
AI-Powered Diagnostics Software Guides Cardiologists during Angiography
Researchers are in the initial stages of a pivotal multi-center clinical trial to demonstrate the efficacy of AutocathFFR, a fully automated system that detects stenoses (narrowing) in the coronary arteries surrounding the heart. AutocathFFR provides cardiologists with the relevant physiological parameters that can help them in assessing the severity of their patients' condition and devise the optimal treatment strategy.
Read More
06 Nov 2019
IoT-Enabled Healthcare Equipment Market Projected to Reach USD 69.7 Billion by 2023
The global IoT-enabled healthcare equipment market is projected to grow at the rate of 29.9% annually from USD 18.8 billion in 2018 to USD 69.7 billion in 2023, driven by digitization during the forecast period. Medical devices companies are transforming from device or consumable providers to disease and care management organizations. Digitization in acute or chronic care is greatly aided by IoT-enabled medical devices, such as subcutaneous drug delivery units, continuous glucose monitors and vitals monitoring equipment. The emergence of digital enterprises such as Google and Apple in the consumer health space has led to development of new opportunities and innovations for medical devices companies.
Read More
05 Nov 2019
AI-Driven Software Uses CT Scans to Determine Extent of Brain Injury in Stroke Patients
A new study has shown that an artificial intelligence (AI) driven software can reliably tell doctors how much brain is injured in stroke patients using universally available CT scans, without the need for more complex advanced imaging.
Read More
Copyright © 2000-2019 TradeMed.com. All rights reserved. | Terms And Conditions