DeepCyto Files IPO Prospectus: Pioneering AI Solutions for Integrated Hematopathology Diagnosis
The Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences, is a world-leading specialized hospital for hematological disorders. Every day, the hospital is crowded with patients from across China seeking treatment for blood diseases. Wang Zhigang, CEO of DeepAnalysis AI and father of a child with leukemia, has personal experience with this reality. He stated, “Due to the challenges in diagnosing blood diseases, the scarcity of primary healthcare resources, and the prolonged course of leukemia treatment, many patients from other regions are forced to seek medical care far from home for extended periods, imposing a substantial disease burden.”
Hematopathology is a relatively independent discipline and represents a highly specialized, cutting-edge service in hospitals. Currently, only more than 800 hospitals across China have the capability to perform hematopathological diagnostics. Moreover,Due to the complex classification of hematologic diseases and the challenges in pathological diagnosis, even physicians at major hospitals face issues of low accuracy and slow efficiency, while hematopathological diagnosis remains virtually nonexistent at the primary care level.China urgently needs to improve the accuracy and efficiency of hematopathology diagnosis and promote the decentralization of high-quality medical resources.
Wang Zhigang believes that,AI technology is the breakthrough for improving the efficiency of hematopathology diagnosis and alleviating the strain on medical resources.This was also the entry point for Wang Zhigang to found DeepAnalysis Intelligence.
Wang Zhigang holds a Bachelor’s degree in Engineering Mechanics and a Master’s degree in Computational Fluid Dynamics from Tsinghua University, as well as a Master’s degree in Computer and Aerospace Engineering from Pennsylvania State University. He has served for many years in senior R&D engineering and management roles at renowned companies such as IBM, Cadence, Mentor Graphics, and Siemens. He also founded the technology company BioCAX, collaborating with the National Institutes of Health (NIH) on the development of medical imaging, human 3D scanning imaging, and smart wearable devices.
In 2016, with the vision of “Achieving Intelligent Comprehensive Diagnosis in Hematopathology to Facilitate the Downward Distribution of High-Quality Medical ResourcesDriven by this vision, Wang Zhigang initiated the development of an AI-based diagnostic product for hematopathology through scientific research collaborations. “At the time, we estimated that the accuracy would reach 70%–80%, but unexpectedly, it exceeded 90%.” Greatly encouraged by this result, Wang returned to China in 2018 to found DeepAnalysis Intelligence, a company dedicated to comprehensive AI-driven diagnostics in hematopathology.Shortly after its establishment, the company secured angel investment from SoftBank China and YuanSheng Venture Capital. In 2019, it completed a Pre-A financing round led by Northern Light Venture Capital.
Leveraging the world’s largest comprehensive MICM database for hematopathology, DeepAnalysis AI has integrated machine learning, deep learning, and big data mining technologies to launch multiple AI solutions tailored for morphological diagnosis of blood cells, flow cytometry diagnosis, and cytogenetic karyotype analysis.Among these products, Deepcell, an AI-based cytological diagnostic solution, can identify more than 40 common cell morphologies associated with diseases such as leukemia and lymphoma. It provides rapid and accurate interpretation of cell morphology with an accuracy rate of up to 97.5%, while reducing interpretation time by 90% year-on-year. DeepFlow, developed by Deepwise Intelligence, is the world’s first AI-powered cloud diagnostic system for flow cytometry. It achieves a diagnostic accuracy of up to 95% for acute leukemia and operates approximately 100 times faster than manual diagnosis. DeepKaryo is China’s first AI-based karyotype analysis system capable of fully automating the entire workflow from chromosome scanning and analysis to report issuance.
Cell morphology is the gold standard for diagnosing hematologic disorders; however, in China, morphological diagnosis remains in a highly “primitive” state, relying on manual assessment. Each case requires manual differential counting of 200–500 cells, involving manual adjustment and counting during the process. This approach is labor-intensive, subject to significant observer bias, and suffers from poor reproducibility and verifiability. Furthermore, due to the lengthy training period required for morphologists—who need years of experience—China is facing a substantial shortage of qualified personnel in cell morphology diagnosis.
In-Depth Analysis of Intelligent DevelopmentThe DCS-1000 is currently the only fully automated cellular morphology microscopic scanning system in China capable of simultaneously processing peripheral blood and bone marrow samples.Capable of automatically scanning and analyzing peripheral blood and bone marrow cells, featuring dual-precision scanning and fully automated slide loading and oil immersion. It completes the scanning of an entire slide within 5 minutes and, when integrated with the DeepCell AI-based cell morphology analysis software, can automatically generate reports.
Shenxi Intelligence is collaborating with the Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences; Peking University First Hospital; KingMed Diagnostics; and Beijing Chaoyang Hospital to conduct clinical validations. Data show that the DCS-1000, when used in conjunction with DeepCell, can differentiate more than 40 common cell morphologies associated with diseases such as leukemia and lymphoma, enabling rapid and accurate interpretation of cell morphology with an accuracy rate of up to 97.5% and reducing interpretation time by 90% year-on-year.
According to Wang Zhigang, the DCS-1000 is about to obtain a Class II medical device registration certificate in China and is preparing to submit a 510(k) premarket notification application to the U.S. FDA. “It is worth noting thatThe DCS-1000 does not require application for new billing codes, enabling rapid market entry upon approval.”
Flow cytometry is a commonly used technique in hematopathology diagnosis and a key focus of DeepAnalysis Intelligence’s strategic layout. Wang Zhigang stated, “Clinical flow cytometry diagnostics are not lacking in instruments and reagents; rather, there is a shortage of professionals with data analysis capabilities. Therefore, Shenxi Intelligent’s AI analysis software will significantly expand the application of flow cytometry in clinical diagnostics.”
Flow cytometry is a high-throughput technology characterized by ease of operation, rapid analysis, and objective precision. However, similar to the challenges faced in morphological diagnosis, flow cytometry data analysis in China still relies heavily on manual interpretation. Manual analysis is subject to significant subjective bias, exhibits low sensitivity, and is prone to missing residual cancer cells at extremely low densities following chemotherapy. Furthermore, manual analysis is time-consuming and labor-intensive, typically requiring more than 10 minutes to half an hour per case. Taking leukemia as an example, immunophenotypic analysis involves examining 30–40 markers across up to millions of cells, resulting in a substantial workload for manual processing. Additionally, the learning curve for mastering manual flow cytometry data analysis is steep, leading to a severe shortage of qualified professionals with flow cytometry analytical capabilities in China.
DeepFlow is the world’s first AI-powered automated diagnostic platform for flow cytometry, developed by DeepAnalysis Intelligence, and is compatible with all mainstream flow cytometers.Provides automated cloud-based management, storage, and diagnostic workflows for flow cytometry data, automatically identifying common hematologic disorders and immune abnormalities with analysis completed in just 5–10 seconds. This not only eliminates subjective errors introduced by manual gating and interpretation but also significantly enhances the efficiency and accuracy of flow cytometry diagnostics.
DeepFlow has already initiated clinical trials at Peking University First Hospital, the Institute of Hematology and Blood Diseases Hospital of the Chinese Academy of Medical Sciences, the First Affiliated Hospital of Sun Yat-sen University, The University of Texas MD Anderson Cancer Center in the United States, and Oregon Health & Science University Hospital. Meanwhile, DeepAnalysis AI has signed a strategic partnership agreement with Cytek, a U.S.-based flow cytometer manufacturer, to provide end-to-end solutions for customers.