알림 목록

Analysis Area: Daegu Metropolitan City

Core Area: Daegu-Gyeongbuk Advanced Medical Complex · Dong-gu Innovation City · Regional University Hospitals · DGIST and Medical Device · Digital Health Corporate Zone

Agenda:  Is the aggregation of hospital, clinical, medical device, and brain disease data and AI companies leading to product approval, commercialization, revenue, and global expansion?

Golden Time Type: Data Commercialization Gap + Regulatory Time Risk

Reference Date: August 28, 2026

Version: Regional AX Golden Time Intelligence v3.2
 

Daegu is home to a concentration of four medical colleges, more than five tertiary general hospitals, approximately 3,800 medical institutions, and the Daegu-Gyeongbuk Advanced Medical Complex . With the addition of medical data, AI medical devices, digital therapeutic devices, and AX for brain diseases to the foundation of a medical service-centered Medicity, the axis of competition has shifted from hospital scale and clinical performance to data-driven product development and commercialization.  However, the ratio of hospital infrastructure to local AI companies' licensing, sales, and exports has not been publicly disclosed; if the utilization of medical data and the adoption of corporate products are delayed over the next two to three years, the existing medical city and the AI ​​healthcare industrial city will become separated. While  Daegu possesses the foundation to expand into AI healthcare, the results of this industrial transformation remain limited.

The combined revenue of the Daegu-Gyeongbuk Advanced Medical Complex and the Medical R&D District in 2021 was approximately 1.8651 trillion won , and verifiable employment stood at 5,804 people as of 2020. While the corporate foundation of the medical industry has already been established, these statistics do not separate AI medical devices, digital healthcare, pharmaceuticals, and general medical devices, and updates to the latest data are limited. Unless AI healthcare revenue and employment are tracked separately until 2028, the growth of existing medical companies will be mixed with AX performance.  Although the industrial scale of Medi-City Daegu can be confirmed, the independent scale of the AI ​​healthcare industry remains undetermined.

Daegu is implementing the "K-Health National Medical AI Service and Industry Ecosystem Establishment Project" from 2023 to 2027, with a total budget of 14.5 billion KRW , investing 10 billion KRW in state funds, 3 billion KRW in municipal funds, and 1.5 billion KRW in private funds. By incorporating medical data collection, processing, and brokerage; AI development; licensing; demonstration; marketing; and workforce training into a single support system, the scope has been expanded from simple medical data construction to full-cycle support for product commercialization. However, cumulative performance metrics linking the number of supported companies, products approved by the Ministry of Food and Drug Safety, regular hospital adoption, and revenue are not clearly verifiable in publicly available data. If product adoption does not occur before the project concludes in 2027, the results of data, consulting, and demonstration cannot be transferred to the commercial market. While the AI ​​medical ecosystem project is currently underway, its industrialization outcomes remain awaiting verification.

In 2026, 14 institutions, including five tertiary general hospitals in Daegu , launched a cooperative structure for the joint utilization of medical data dispersed across hospitals and the linkage of review procedures. While the structure has shifted from data utilization at the individual hospital level to the linkage of medical data at the city level,  common data models, federated learning, institutional responsibilities, corporate access times, and utilization costs have not yet been confirmed. If the agreement does not transition to actual datasets and multi-institutional verification within two to three years, companies will have to repeatedly enter into separate IRBs and DRBs for each hospital.  Although the joint utilization of medical data has begun, it is too early to determine whether it constitutes an operational data network.

Daegu's Digital Healthcare Medical Device Demonstration Support Project proposed forming medical institution-company consortia for products approved by the Ministry of Food and Drug Safety in 2026, along with pilot distribution to five medical institutions and support of up to 180 million won . Although the scope of support has shifted from clinical planning to actual pilot distribution to hospitals , post-support purchases by medical institutions, insured and non-reimbursable revenue, and repeat usage rates remain unconfirmed. If no regular purchases without subsidies occur by 2028, hospitals will remain as demonstration sites and companies as project implementers. While  Daegu has initiated the field entry of AI medical devices, the formation of a self-sustaining hospital market remains unconfirmed.

The K-MEDI hub supports new drug development, advanced medical devices, preclinical trials, accredited testing, and pharmaceutical production, and possesses infrastructure for MRI, bioimaging, digital pathology, and medical device evaluation. While the infrastructure for research, testing, and verification of medical products has been established, AI medical devices require verification of data quality, algorithm performance, and real-world clinical environments, as well as continuous model management, resulting in an operational structure different from that of existing medical device support.  A regional performance management system for re-verifying models after updates and tracking liability for bias, performance degradation, and medical accidents has not been identified. If lifecycle management for AI products is not established by 2028, hospital expansion will be halted following approval and initial demonstration.  Although Daegu has a strong foundation for medical devices, its operational infrastructure for AI medical devices is in its early stages.

The Advanced Medical Complex is occupied primarily by R&D institutions and corporate research laboratories, with general manufacturing restricted. While suitable for linking research, preclinical, and clinical trials, the structure requires mass production and large-scale employment to be relocated to production facilities outside the complex . Although AI medical devices have low dependence on manufacturing plants, the dispersion of software companies across Suseong Alpha City necessitates transaction outcomes between the medical complex, hospitals, and digital firms. If joint products and contracts are not established between the spaces within two to three years, the industrial value chain will become fragmented, even if individual infrastructures remain strong.  While Daegu possesses numerous medical and digital assets, its operation as an integrated industrial hub remains incomplete.

Daegu possesses the infrastructure for secure medical data utilization at Keimyung University Dongsan Hospital and a multi-institutional medical data cooperation system, and its experience in building datasets for neurological diseases, such as cerebral infarction MRI and clinical data, has also been confirmed. While the foundation for shifting data from individual research assets to assets for AI medical device development has expanded, differences in format, quality, labeling, and review procedures among hospitals remain.  The time taken from a company's application to data access, approval rates, and the number of multi-institutional data utilization cases are not disclosed. If access times remain longer than the product development cycle until 2028, Daegu's data holdings will not translate into corporate competitiveness.  Medical data assets exist, but they remain incomplete as marketable industrial assets.

The 2026 multi-institutional agreement presented a direction to link complex review procedures and expand the joint use of medical data. However, centralized data integration faces significant constraints due to patient personal information and institutional responsibilities, and  evidence of the actual operation of federated learning or distributed analytics structures has not been confirmed. Unless technical and legal operational models are finalized within two to three years, joint use will remain at the level of approval and cooperation from individual hospitals.  Daegu's medical data governance is currently transitioning from the agreement stage to the implementation stage.

Daegu possesses the demonstration of digital medical devices for brain and developmental disorders, the utilization of cerebral infarction MRI and clinical data, and the evaluation infrastructure for brain disease models and bioimaging through the K-MEDI hub. The Regional Hub AX Innovation Technology Development Project has also included brain disease diagnosis and treatment, along with AI-based digital medical products, as specialized axes for Daegu, thereby expanding individual research into a regional strategy. However, AI products for brain diseases that link local companies with KFDA approval, regular adoption by medical institutions, and patient usage to revenue are confirmed only on a limited basis. Unless the first commercial reference is established by 2028, this specialization will remain merely a research project title.  While Daegu has the foundation for AI in brain diseases, specialization in the product market remains unconfirmed.

Kyungpook National University possesses a research infrastructure that combines brain surgery robots, imaging, and electronic engineering with clinical medicine, while the K-MEDI hub provides non-clinical support, including MRI and behavioral assessments. Although the expertise among research institutions is high, a structure in which a single responsible company leads the entire process—from hospital clinical demand to corporate product design, regulatory approval, and sales—is not observed. If research institution-centered consortia are repeated for two to three years, technology may be accumulated, but local companies with product IP and sales capabilities fail to grow.  The supply of research specialized in brain diseases is at a mid-to-high level, but the entities responsible for commercialization remain weak.

From 2024 to 2026, the demonstration support for digital healthcare medical devices continued with clinical planning and pilot distribution; however, for 2026, conditions were set requiring consortia between medical institutions and companies, as well as corporate self-financing. While corporate responsibility for the demonstration has been strengthened, no structure has been identified that pre-binds hospital purchases after the support ends.  Even if clinical efficacy is confirmed, hospitals may choose not to continue using the product if there is no budget, reimbursement, or operational alignment. If the conversion rate to regular purchase remains low until 2028, hospitals in Daegu will be permanently fixed as testbeds where demonstration projects are repeated.  While clinical accessibility is confirmed, market accessibility remains a separate gap.

AI medical devices and digital therapeutic devices must be integrated into medical staff workflows, electronic medical records, patient management, and reimbursement systems for repeated use. While Daegu’s demonstration announcement supports product verification and pilot deployment, data on total operating costs—including hospital system integration fees, additional workload for medical staff, maintenance costs, and liability distribution—is not available. If operational economic feasibility is not verified while product performance is confirmed over a period of two to three years, hospital expansion will be limited.  Daegu’s demonstration framework is centered on technology verification, and the verification of hospital operating models is still in its early stages.

Although Daegu is densely populated with medical institutions and public support facilities, there are no up-to-date statistics separately compiling the number of local AI medical companies, investment attraction, licensed products, sales, and exports.  While infrastructure performance is reflected in facility usage and support volumes, corporate value, recurring revenue, and follow-up investment remain separate. Unless corporate-level performance is accumulated by 2028, AI healthcare anchor companies will not emerge, even if the MediCity brand is strengthened.  Daegu's AI medical ecosystem is institution-led and has not yet reached a stage of enterprise-led growth.

The K-Health project supports overseas field trials, opening a pathway for local companies to secure international references beyond domestic hospitals. However, cumulative results following these trials—leading to local licensing, distribution contracts, and paid sales—are confirmed only to a limited extent. If overseas trials remain at the level of exhibitions and verification for two to three years, companies must independently bear the burden of country-specific regulations and the costs of establishing sales networks. While  Daegu's AI healthcare has begun its entry into the global market, its transformation into an export industry remains unconfirmed.

Since its launch in 2009, Medicity Daegu has built a medical service brand and fostered cooperation among medical institutions, while also operating medical community consultative bodies and medical tourism support systems. In 2025, the AI ​​Bio-Medicity Council was established, expanding the structure to encompass medical services, biotechnology, and industry. However, the contribution of patient care quality, medical tourists, and hospital reputation to revenue for local AI medical companies remains unconfirmed. If performance indicators for medical services and industrial policies are separated by 2028, a dual structure will persist where the Medicity brand is strong but the AI ​​healthcare industry is weak. While  Daegu has established itself as a medical city, it is currently transitioning into an AI medical industry city.

Although Daegu possesses numerous tertiary hospitals and medical schools, concerns are simultaneously being raised regarding the outflow of patients to hospitals in the capital area and the burden on essential local medical services. In a situation where medical institutions are under pressure regarding treatment, personnel, and finances, corporate demonstration and data provision can act as additional workloads. Unless the costs of industrial cooperation for hospitals are separately measured over a period of two to three years, data and clinical collaboration rely on the participation of individual medical staff.  The structure in which medical demand and industrial demonstration compete for the same hospital resources is identified as a constraint on the expansion of AI healthcare in Daegu.

Tertiary general hospitals, medical colleges, advanced medical complexes, the K-MEDI hub, and the K-Health project coexist. While  the support infrastructure for clinical, non-clinical, data, and licensing is at a top-tier domestic level, licensed products, hospital purchases, and recurring revenue from local AI medical companies are observed only on a limited basis. If institutional capabilities are not transferred to corporate performance by 2028, a structure dependent on public support will become entrenched. Currently, the level of readiness is assessed as high for medical and research infrastructure and medium-low for corporate commercialization .

The 2026 demonstration support was scheduled for pilot deployment at five medical institutions and stipulated a medical institution-company consortium as a condition. While on-site entry into hospitals has expanded, the regular adoption rate, usage rate, and purchase amount after the support ends are not verified. If only new demonstrations are accumulated over two to three years, the increase in the number of hospitals overrepresents the expansion of the commercial market. The level of hospital expansion is determined by the progress of pilot adoption and unconfirmed regular purchases .

There are numerous companies with sales exceeding 10 billion won in the Advanced Medical Complex and Medical R&D District, and the overall revenue base of medical companies is also confirmed. However, there are no recent statistics separating the number and revenue of AI medical device, digital therapeutic, and medical data companies. Unless existing medical companies and AI companies are distinguished by 2028, it is impossible to determine whether the industry will expand. The level of corporate diffusion is assessed as having a moderate to high base of medical companies and unconfirmed diffusion of AI healthcare companies .

Experience in establishing multi-institutional medical data agreements, a foundation for safe utilization, and brain disease imaging data infrastructure is confirmed. On the other hand, data standards, approval periods, and corporate access rates by hospital, as well as the regional retention scale of clinical-AI convergence personnel, were not disclosed. If data utilization relies on institutional reviews and a small number of researchers for two to three years, the speed of corporate development will not increase. The readiness level for data and talent is assessed as upper-middle for assets held and lower-middle for industrial accessibility .

The K-Health project is scheduled to end in 2027, and the Digital Healthcare Medical Device Demonstration Support in 2026, while the Regional Hub AX project continues until 2030.  An overlapping structure has been formed where larger AX investments in brain diseases begin before the product and corporate performance of existing demonstration projects is verified. If hospital purchases and revenue conversion rates from previous projects are not confirmed by 2028, the new projects will repeat the same demonstration-termination pattern. The golden time for the transition to AI healthcare in Medicity Daegu is assessed to be the next 24 months .

While medical data agreements, demonstration centers, and research equipment enable long-term operation, AI companies are highly likely to relocate to hospitals in the metropolitan area and capital markets if delays in licensing, clinical trials, and investment accumulate.  Once hospitals experience demonstration fatigue and a loss of specialized personnel, the cost of regaining participation from companies and medical staff increases significantly, even if new projects are introduced. If commercial products and purchasing markets are not established by 2028, Daegu will enter a structure where it retains only medical research and testing infrastructure, while corporate growth is driven out of the region . Irreversibility is assessed as medium risk for infrastructure, high risk for corporate outflow, and medium-to-medium risk for data networks .

Medicity Daegu has expanded beyond medical service brands to include medical data, AI medical devices, digital healthcare, and AX for brain diseases.  Its research, hospital, and demonstration infrastructure is at a level ready to enter actual industrial transformation.

However, industrial performance is still concentrated in the public projects, agreements, and demonstration stages . The corporate growth trajectory leading to licensed products, regular hospital purchases, repeat sales, and global sales has not been sufficiently confirmed.

The final judgment is “Clinical Infrastructure Strength + AI Healthcare Commercialization Gap .  ” Medicity Daegu is expanding into the AI ​​healthcare industry, but it did not reach a judgment that it has transitioned into an independent growth industry.

Overall Score: 60 / 100

Golden Time Status:
ORANGE

Time Window:
Approx. 24 months

Failure Mode:
Hospitals, data, and research facilities remain in Daegu, but the revenue, investment, and growth of AI medical companies shift to the Seoul metropolitan area.

Irreversibility:
High risk of corporate leakage and repeated validation

Evidence Sources

• Medicity Daegu—Based on medical institutions, universities, and the medical industry
• Daegu Medical Enterprise Support Portal—Sales and Employment in the Advanced Medical Complex and Medical R&D District
• Ministry of Science and ICT—K-Health National Healthcare AI Services and Industrial Ecosystem
• Korea Information & Communication Technology Promotion Agency—Daegu AI Medical Ecosystem Establishment Project
• Daegu Technopark—2026 K-Health Daegu AI Medical Ecosystem
• K-MEDI hub—Pilot deployment of digital healthcare medical devices in 2026
• K-MEDI hub—Digital healthcare medical device demonstration support project
• Daegu Metropolitan City Council—AI Brain Developmental Disorders and Medical Data Project Subject to Audit
• K-MEDI Hub—Support System for New Drugs, Medical Devices, Non-clinical Research, and Pharmaceutical Manufacturing

Structural Insight — The success or failure of medical data sharing hinges on 'review time integration' rather than data integration

Even if large hospitals possess sufficient medical data, companies must separately pass the IRB, DRB, security reviews, and contracts of each hospital. Verifying the same AI medical device across multiple hospitals requires repetitive similar reviews and data maintenance, and  administrative procedures can take longer than the actual data analysis.

Consolidating all medical data in one place is difficult to implement due to issues regarding personal information, liability, and security. On the other hand, if hospitals retain their data while applying common application forms, evaluation criteria, and analysis environments, multi-agency verification time can be reduced without physical integration.  In this structure, a city's competitive asset is not the total amount of data, but the time saved when a company enters a second hospital.

Therefore, the final outcome of the Daegu medical data network is not judged solely by the number of participating hospitals or data records. Whether verification that took 12 months at the first hospital is shortened to 3 months at the next hospital determines the actual speed of the AI ​​medical industry. The data competitiveness of Medicity Daegu is judged not by how much data has been collected, but by how much the time required for multi-institutional verification repetition has been eliminated .