A Software Lets Doctors Build Deep Learning Models Themselves.

The COVID-19 pandemic is radically changing all parts of the industries. The medical industry is no exception. The unprecedented challenges posed by the pandemic is forcing the industry to make best use of scarce medical resources and make medical care more efficient by adopting innovative technologies.

A deep learning model can be created by medical staff to analyze X-ray photos themselves and to be used for diagnostic assistance/Photography provided=IBK blue air Mapo

Among various technologies that are applicable to the medical industry, deep learning vision technology (image interpretation technology using deep learning), is particularly receiving a lot of attention. The frequent use of CT, MRI, and X-ray scans in the medical sector has always meant the need for advanced image analysis. Deep learning is heralded as the gold standard of image analysis given its ability to learn without human supervision.

Neurocle, an AI startup specializing in deep learning vision, has garnered much attention by releasing Neuro-T & Neuro-R which is general-purpose software that allow the medical industry to meet it unique demands better than traditional technology.

Deep learning model with limited resources

There are already a variety of AI-based medical diagnostic solutions companies. But what they focus on is specific disease-oriented solutions. In order to build such a solution, it is common for large hospitals and AI companies to sign MOUs and provide customized diagnostic services with hundreds of thousands of pieces of data, a large number of people, and costs. However, it is very difficult to find suitable AI-based medical diagnostic solutions provided by the market at an early stage of research on a few or new diseases with little data. However, with Neurocle's deep learning vision software, anyone can easily create customized medical image interpretation deep learning vision models.

Deep learning model without deep learning experts

If a business possesses plentiful images of scans but lack deep learning experts, this poses another barrier to building and implementing a deep learning model for medical research or diagnosis. Neurocle’s software that leverages Auto Deep Learning technology help eliminate the need for deep learning experts. When enough images are uploaded to the software, Auto Deep Learning allows it to build a deep learning model without any coding. Because Auto Deep Learning chooses the optimal set of hyperparameters for the user, it may be particularly useful for medical personnel conducting actual research.

Deep learning and information security

Despite the promises of image analysis through the web- or cloud-based solutions, the concern over security and privacy of medical information cannot be easily removed. However, these risks can be mitigated if users can create and deploy deep learning models from their own local PC. Neurocle provides On-Premise software (On-Premise: a method to directly install software server in local PC). This is most welcomed for businesses in the medical sector, a lot of which handle sensitive, confidential information.

"We have made efforts to develop deep learning vision software that can be used by non-experts so that deep learning can be freely applied to various disease images. Deep learning vision technology is no longer difficult, so it will be used more diversely in the medical field." In fact, it is said that Neurocle's Deep Learning Vision software are widely used to apply deep learning vision technology to fields that are being researched at domestic leading university hospitals. Neurocle is a deep learning vision software company that has attracted attention from a wide range of industries by providing standard software products (Neuro-T & Neuro-R) that allow anyone to create deep learning models and classify images regardless of user expertise. In addition, it completed partnership contracts with global video data companies and succeeded in entering about 10 countries. In January this year, it was selected as the 4th IBK Changgong Mapo Co., Ltd., a startup development platform run by IBK Industrial Bank.

Source: https://news.mt.co.kr/mtview.php?no=2020060117532226485