Anhui Medical's Huan Dandan: Intelligent Capsule Endoscopy Robot Enables Early and Precise Screening of Gastrointestinal Diseases | Forum Speech

Healthcare has a strong potential demand for artificial intelligence. Currently, a relatively complete industrial structure of “foundation + technology + application” in the global AI-plus-healthcare sector has initially taken shape. For new technologies to truly drive industry transformation, it requires not only coordinated efforts across policy, technology, talent, and other areas, but also corporate exploration and the accumulation of time. To explore future development and practical implementation strategies for health and medical big data and artificial intelligence, the 2017 Yangtze River Industry Forum (Autumn) and the Health and Medical Big Data & Artificial Intelligence Conference were grandly held at the Wuhan Convention Center on September 16–17, 2017.

At this conference, Huan Dandan, Vice President of Anhan Photoelectric Technology (Wuhan) Co., Ltd.,“Smart Capsule Endoscopy Robots Facilitate Early and Precise Screening for Gastrointestinal Diseases”Titled “...”, the speech elaborated on AnHan’s current explorations in AI-assisted diagnosis based on capsule gastroscopy, as well as the company’s future plans. Below are the highlights of the speech compiled by VCBeat:

Guest Introduction

With the advancement of medical technology, endoscopes have become an extension of physicians’ hands and eyes. AnHan Medical is a company specializing in the detection of gastrointestinal diseases using endoscopic techniques. What I have brought to the venue today is a compact capsule robot. Approximately 27 millimeters in length, it can move with millimeter-level precision under magnetic field control, enabling comprehensive, precise, 360-degree blind-spot-free examinations within the stomach.

As is well known, China bears the highest global burden of gastric cancer, with both its incidence and mortality rates ranking first worldwide. In China, nearly one person dies from gastric cancer every minute. According to statistical data from 2015, there were 680,000 new cases and 500,000 deaths annually in China. Although gastric cancer is formidable, if detected at an early stage, it can be treated via endoscopic submucosal dissection (ESD) through the mouth without the need for open surgery, achieving a five-year survival rate of over 90%. However, the reality in China is that 90% of patients are diagnosed at middle or advanced stages, resulting in a five-year survival rate of less than 20%.

The primary reason why large-scale gastric cancer screening cannot be widely implemented in China is that early-stage gastrointestinal diseases often present with nonspecific symptoms, and traditional intubation-based gastroscopy is painful. Therefore, there is a need for a highly comfortable diagnostic method to enable early diagnosis of gastrointestinal diseases and early detection of tumors. AnHan’s capsule robot offers multiple capabilities, transforming a tube over one meter in length into a small capsule held in the hand that can enter the human body. Under the direction of physicians, it can accurately perform comprehensive observations of the gastrointestinal tract.

Patients need only swallow the capsule to complete a gastroscopy, offering a painless and non-invasive procedure without the need for anesthesia or risk of cross-infection. Although small in size, the Anhan Capsule Gastroscope integrates more than 300 components, delivers exceptionally clear images, and can operate continuously for over 10 hours.

If such comfortable diagnostic methods become widely adopted, the volume of image interpretation in hospital gastroenterology departments will inevitably surge. If a hospital’s annual capsule endoscopy usage reaches 2 million cases, with each case requiring the capture of 3,000 images, physicians would need to review a total of 6 billion images. Medical resources in China are unevenly distributed and severely limited. Each year, 80 million to 100 million people in China require gastroscopy; however, there are currently only more than 26,000 specialized physicians nationwide qualified to perform gastroscopies, making it impossible to conduct large-scale early screening for digestive tract diseases.

Therefore, we need to leverage the Internet to facilitate data transmission and sharing.First, by leveraging gastrointestinal imaging acquisition technologies and the “Internet+” model, we implement tiered healthcare delivery, separate examination from diagnosis, and transform patient mobility into data flow. Then, using cloud platform technology, we connect physicians, patients undergoing examinations, and diagnostic institutions to achieve resource sharing. Finally, by applying big data and artificial intelligence technologies, we reduce the cost of image interpretation and realize smart healthcare.

The cloud platform addresses challenges in data storage, sharing, and management, enabling image-based diagnosis and online multi-specialist consultations anytime and anywhere with internet access. By leveraging the internet, it enhances physicians’ work efficiency. Furthermore, we have integrated resources on the platform, with all connected community hospitals, health examination centers, Grade A tertiary hospitals, and international medical institutions onboarded. Consequently, it is highly feasible to engage experts from international medical institutions to provide diagnostic analysis for patients.

Currently, more than 100 of China’s top-tier hospitals, including Wuhan Union Hospital, have clinically adopted the Anhan Capsule Gastroscopy. Additionally, four to five hundred health examination centers are also using our product.

On April 20 this year, the Anhan Intelligent Capsule Endoscopy Robot Imaging Diagnosis Platform was integrated with IBM’s medical imaging cognitive analysis technology, aiming to assist gastroenterologists in making more accurate and rapid clinical judgments. The capsule robot AI-assisted diagnostic system we developed in collaboration with IBM leverages deep learning on an artificial intelligence platform to establish a diagnostic and disease screening platform for the gastrointestinal tract.

This system employs techniques such as data augmentation, focused learning, and transfer learning to construct deep learning networks, currently achieving an accuracy rate of over 90%. By leveraging artificial intelligence to assist in diagnosis, it significantly reduces the volume of images physicians need to review. The workload has been reduced from more than 20,000 images to merely dozens for physician review.

In the future, we aim to cover the entire spectrum of gastrointestinal medical care, including surgical assistance capsules and capsules for targeted drug delivery. Our capsule robots are now capable of precise control and positioning. Building on this platform, we intend to expand into all aspects of gastrointestinal diagnosis and treatment, while integrating all our products with big data and artificial intelligence platforms.

Furthermore, the vibrating capsule has already entered clinical trials. This capsule extends into the therapeutic domain and is similar in size to a capsule endoscope. It is essentially an electronic drug that does not require administration by healthcare institutions; patients can use it at home, controlling its vibration frequency and intensity via mobile devices to massage the gastrointestinal tract, thereby replacing chemical medications for the treatment of conditions such as constipation.

Finally, we also hope to bring more advanced and comfortable medical experiences to the general public through big data and artificial intelligence, enabling Anhan Intelligent Manufacturing to usher in a “Healthy Gastric Future.”