Samsung to participate in the North American Society of Radiology (NARS) from the 25th to the 30th.
Exhibition and symposium on medical imaging equipment lineup
AI algorithm-based diagnostic assistance function introduced
Samsung Electronics and Samsung Medison participated in the 'Radiological Society of North America 2018 (RSNA)' held in Chicago, USA from the 25th to the 30th, and unveiled their entire line of imaging diagnostic devices, including ultrasound, digital X-ray, CT (computed tomography), and MRI (magnetic resonance imaging), and showcased Samsung's unique AI diagnostic assistance functions.

The need for AI-based diagnostic devices continues to grow, both to support hospitals and medical staff in developing countries with relatively few doctors per thousand people and to reduce misdiagnosis rates for difficult-to-diagnose lesions.

Considering this medical environment, Samsung is actively applying AI technology, focusing on highly useful functions in each product group, and at this event, it provided an opportunity for academic experts to intensively experience AI-based diagnostic assistance functions.
The 'Ultrasound Zone', which showcased premium ultrasound equipment for radiology, highlighted the 'S-Detect for Breast' function, which analyzes the characteristics of selected lesions in breast ultrasound images using AI technology.

This function improves the accuracy of breast cancer diagnosis by unskilled medical staff by learning big data consisting of approximately 10,000 ultrasound diagnosis cases, helping to determine the presence or absence of lesions, analyzing their characteristics, and providing a standardized diagnosis report.

According to a paper published this year by Professor Tommaso Vincenzo Bartolotta, an Italian radiology expert, using the 'S-Detect Forbreast' function improved diagnostic accuracy from 0.93 to 0.95 on a scale of 1 to 1 for specialists with more than 10 years of experience, and from 0.83 to 0.87 for those with around 4 years of experience, proving that this function is particularly useful in hospitals with a shortage of skilled medical staff.

In 'X-ray Zone', functions that apply AI techniques in image processing were introduced. Representative examples include the 'Bone Suppression' function, which removes the ribs from chest X-ray images to clearly show lung lesions hidden by bones, and 'SimGrid', which provides a similar level of image quality while lowering the radiation dose by one level without the grid, an auxiliary component used for clear images.

According to research results presented by Professor Hong Gil-seon of Seoul Asan Medical Center at the Radiological Society of North America in 2017, the 'Bone Suppression' function helps inexperienced readers interpret difficult lung lesions such as pneumonia, tuberculosis, and lung metastases.

ALND (Auto Lung Nodule Detection), a lung nodule diagnostic assistance function currently under review by the U.S. Food and Drug Administration (FDA), has attracted attention as an AI CAD (Computer Aided Detection) solution based on a deep learning algorithm.

According to research results announced this year by Professor Myung-Jin Jeong of Samsung Seoul Hospital, ALND's detection rate for lung cancer smaller than 3 cm was 92%, which was an average of 7 percentage points higher than when diagnosed by a specialist without using this function.
The 'CT Zone' exhibited a mobile CT equipped with a built-in battery that is widely used in various places such as a lung disease CT screening vehicle, a stroke emergency vehicle, intensive care unit, and operating room, and applied AI technology to diagnose stroke patients. The diagnostic assistance function for cerebral hemorrhage that helps in treatment decision-making has been highlighted.

In the 'MRI Zone', a prototype for limb imaging was showcased. It is equipped with a diagnostic assistance function that uses AI technology to compare and learn MRI images of normal people and patients with osteoarthritis, and provides segmented image information on the major tissues of the knee joint in 3D, helping medical staff to more quickly and accurately evaluate the degree of damage to articular cartilage.

In addition to exhibiting diagnostic devices for each product group, Samsung also set up a separate 'AI Zone' so that conference attendees could experience a variety of AI-based diagnostic assistance functions equipped in each product group. It also held a symposium to introduce the core capabilities of Samsung's imaging diagnostic devices that increase diagnostic accuracy and provide a safe treatment environment.

“We are pleased that Samsung’s AI-based diagnostic assistance features applied to existing imaging diagnostic devices are receiving positive responses in the market,” said Dong-soo Jeon, President and CEO of Samsung Medison and Head of the Medical Device Business Division at Samsung Electronics. “As a comprehensive imaging diagnostic device company, we will continue to work closely with hospitals and medical professionals to improve diagnostic accuracy with more advanced AI technology.”