[E-Daily Reporter Hong Ju-yeon ] "We are a company that uses robotics and artificial intelligence (AI) to address the reality where cancer goes undetected due to a shortage of personnel."

Ahn Seong-wook, CEO of RINORBIT, introduced the company in this way. RINORBIT is an AI medical device company that develops integrated AI digital pathology solutions. The integrated AI digital pathology solution consists of the “Bioscope”—an AI scanner (Robotic AI Whole Slide Scanner) that automatically digitizes pathology slides—AI diagnostic software that analyzes cells, and an AI pathology system responsible for data storage and management. CEO Ahn, a visually impaired entrepreneur and graduate of Seoul National University’s Department of Chemical Engineering, said, “I will use technology to resolve the structural contradictions in pathological cell diagnosis, which is directly linked to life.”

Founded in 2023, Rinobit reached the mass production stage just three years after its inception. This achievement was the result of extensive preparation. CEO Ahn monitored the necessary technologies starting in 2016, fully conceptualized the business, and then put his plan into action. “I discarded ideas that wouldn’t work through mental experiments and only proceeded with those that would,” he explained, detailing how he was able to move forward quickly without trial and error.

CEO Ahn Seong-wook, born in 1973, lost his vision to glaucoma seven years ago. However, rather than giving in to despair, he launched his business driven by a conviction to address healthcare inequality. He chose cervical cancer as his first target because it is the type of cancer with the widest disparity in access to medical care. While the cure rate for cervical cancer exceeds 95% when detected early, prevalence and mortality rates remain high in developing countries due to a shortage of pathologists, which leads to delayed detection.

CEO Ahn said, “In developed countries, it can serve as a tool to assist with diagnosis, while in developing countries, it can replace human labor.” The vision is that once digitization is achieved, remote screening will become possible, allowing South Korea’s medical infrastructure and personnel to be extended to developing countries. In fact, discussions regarding exports to developing countries are already underway at the NGO and Official Development Assistance (ODA) levels.

Utilizing AI significantly boosts diagnostic productivity. According to Rinobit, while a single pathologist previously reviewed 120 cases per day, a single device can process up to 180 cases, increasing productivity by 1.75 times and reducing the cost per case by 43%. It can even reduce errors in visual diagnosis. However, unlike in radiology, AI in the field of pathology has not yet taken off. This is because the delay in digital transformation has prevented the acquisition of vast amounts of training data. This is why Rinobit has stepped in to fill this gap.

A Growing Market, a Shortage of Personnel

The cervical cancer diagnosis market is projected to reach approximately 20 trillion won by 2033. In particular, the Asia-Pacific market is growing at an annual rate of 8%. Cervical cancer screening tests are performed 300 million times annually worldwide, with approximately 8 million conducted in South Korea alone. However, the replacement rate for pathologists responsible for interpreting these results hovers in the 10% range, resulting in an acute shortage of personnel.

This is also why the adoption of AI in pathology has been slow. While radiology departments completed the digitization of MRI and CT scans about a decade ago, enabling AI training with vast amounts of data, pathology departments still examine slides under a microscope. The delay in digital transformation has also delayed the adoption of AI.

The reason lies in the sheer volume of data. CEO Ahn explained, “Digitizing a single slide generates about 3 GB (gigabytes),” adding, “It’s on the same scale as viewing a single car on the ground from a satellite.” While this was difficult for hospitals to manage with past technology, the situation has changed as storage device prices have fallen and computer performance has advanced.

Differentiation Through On-Device AI… Conquering ‘Asymmetric Data’

Renobit’s competitive edge lies in its simultaneous development of both hardware and software. While other companies process data from equipment after the fact, Renobit employs an “on-device AI” architecture where the AI runs directly on the hardware to control the equipment and perform analysis. CEO Ahn explained, “Just as a person observing under a microscope would zoom in on a suspicious area to examine it more closely upon noticing something unusual, AI can also perform additional, precise analysis of that area when it detects an abnormal signal.” He added, “By reducing unnecessary analysis and focusing on what’s needed, inspection time is shortened and accuracy is improved.”

Technological precision is another strength. Rinobit analyzes a single cell at a resolution of over 13,000 pixels. The company has also independently developed “focus stacking” (Z-stacking) technology, which combines images taken at multiple focal points to interpret the three-dimensional structure of cell clusters. This method uses AI to merge images from multiple layers—each taken at slightly different depths along the Z-axis of the slide—into a single image file.

The company also cited its workforce composition as a strength. The company achieves a “harmony between experience and innovation” by having professionals with over 25 years of experience handle hardware requiring precision maintenance, while emerging talent manages the AI. They also emphasized that they are not tied to any specific equipment manufacturer. CEO Ahn stated, “AI requires fine-tuning based on equipment characteristics, so vendor lock-in could become a problem later on,” adding, “Rinobit is free from this dependency.”

Renobit recently received the Best Paper Award at “AiMH 2026,” an international AI medical conference held in Palma de Mallorca, Spain. CEO Ahn remarked, “We received this award because the technology has commercial value.” He explained that the technology was recognized for its ability to boost AI performance and reduce costs without incurring additional expenses.

The core of the award-winning technology lies in solving the “asymmetric data” problem. In medical data, there are many normal cases but few abnormal ones. AI learns more easily when normal and abnormal data are similar, but this is not the case in reality. Rinobit proposed a technique to improve accuracy without significantly expanding the dataset. This reduces training costs and allows for improved AI performance without changing the model or existing equipment.

Data is the foundation of its competitiveness. Rinobit has secured a clean dataset of 58,000 cases that have been tagged by medical specialists. The company explains that this high-quality data ensures high diagnostic accuracy and creates a barrier to entry for latecomers.

Domestic shipments to begin in the second half of the year; 2030 revenue projected at 40.3 billion won… Targeting the Reimbursed Market

Renobit is targeting the

reimbursed

market, where standard testing methods (PAP and LBC) are covered by insurance. This means entering the regulated market, rather than the non-standard or non-reimbursed sectors. According to the company, Bioscope increases diagnostic productivity by 1.75 times and boosts daily throughput from 120 cases per person to 180 cases per device. The processing cost per case is reduced by 43%, from 2,167 won to 1,235 won.

The commercialization roadmap is also well-defined. The AI scanner completed certification by the Korea Food and Drug Administration (K-FDA) last year and holds ISO 13485 certification and a manufacturing license for in vitro diagnostic medical devices. The AI diagnostic software is undergoing clinical performance testing, with the goal of securing approval by the end of this year. The company plans to rapidly enter the market through a “hardware-first deployment, remote software installation” strategy.

Domestic shipments will begin in the second half of this year. The company has established a revenue base for this year through letters of intent (LOIs) from major hospitals and supply agreements with large contract manufacturers. Clinical trials are currently under discussion with Chung-Ang University Hospital; as these are retrospective trials using slides from existing tests, the company expects them to be completed within three months at the earliest. CEO Ahn stated, “We will easily achieve this year’s target of 1.5 billion won in equipment sales alone.”

The company also presented its mid- to long-term outlook. Rinobit aims to turn a profit by 2028 and achieve 40.3 billion won in revenue and 12.7 billion won in operating profit (an operating profit margin of 37.8%) by 2030. The company plans to drive profitability by increasing the share of AI software revenue from 39% in 2028 to 63% in 2030. To date, the company has raised 4 billion won in funding and is currently conducting a Series A round; funds from this round will be invested in securing global certifications, data, and key talent. The company plans to begin preparations for an IPO starting in 2029, when revenue is expected to take off.

The company is also expanding its overseas sales network. In the U.S., it has established a joint venture with SP Magna, a medtech specialist, and is preparing for an R&D pilot with a major pharmaceutical company in 2027. CEO Ahn is paying close attention to the trend of major pharmaceutical companies—such as AstraZeneca and Roche—acquiring proprietary data and pathology AI through mergers and acquisitions (M&A).

Rinobit plans to expand its business scope starting with cervical cancer. By using the cervical cancer model as a “fine-source model”—which can be applied to other fields with only minor adjustments—the company aims to rapidly scale up while addressing data scarcity issues.

Urine cytology testing is the next target. Since it can be performed using urine collected during routine health screenings without the need for separate samples, it offers high patient convenience. CEO Ahn explained that this could lead to the diagnosis of urethral and bladder cancers. Diagnosis of thyroid and animal cells is also on the agenda for future expansion. He noted that demand for veterinary diagnostics is growing due to the expansion of the pet market, and since these services are not covered by insurance, they are also advantageous for veterinary clinics’ profitability.

Regarding Rinobit’s goals, CEO Ahn stated, “In the short term, we aim to become a company that accurately diagnoses cervical cancer; in the long term, we aim to become a company capable of diagnosing all cancers through pathology. Ultimately, we will become a company that resolves the issue of healthcare inequality caused by a lack of personnel, which prevents the early detection of cancer.”