Meet the DHA Projects
Get to know some of the names, stories, and journeys of DHA program participants.
Die Seite ist aktuell nur auf Englisch verfügbar.
3D Histopath – Bringing Histopathology from 2D to 3D
Dr. rer. nat. René Hägerling
Institute of Medical Genetics and Human Genetics, Charité – Universitätsmedizin Berlin
No cancer diagnosis without histopathology. Histopathology refers to the preparation and examination of tissue samples in order to study symptoms of a disease. In clinical medicine, a biopsy or a surgical specimen is obtained, sliced, and transferred onto a glass slide, stained with a chemical agent and then examined under the microscope by a pathologist. However, today’s histopathology process is restrained by several limitations. Cutting samples into slices may miss relevant locations, distort and damage tissue, and thus often prevents secondary sample analyses, e.g., via genetic sequencing. Current staining agents only penetrate tissue samples relatively slowly. Information on 3D structures such as blood or lymphatic vessels cannot be seen entirely, and pathologists need to examine the sample via “eyeballing,” a process requiring a very high level of specialization which is also exposed to inter-observer variability. In sum, 2D pathology is not fully adequate for handling the information of our 3D human body. 3D Histopath can address this need by developing an end-to-end histopathology pipeline for improved diagnoses and therapy decisions. The 3D Histopath solution consists of two core components: a staining solution and a software solution. The staining solution includes a new staining technology using Nanobodies able to penetrate large tissue samples much faster than traditional staining agents. The software component in development entails a visualization functionality for pathologists to see 3D structures such as vessels, and an indication-specific AI-based analysis functionality to highlight key sample areas, e.g., metastasizing regions. Initial use cases include lymphedema, lipoedema, and cancer. In the future, other use cases will be implemented. Based on these benefits, 3D Histopath aims to improve and speed up the clinical histopathology process for better diagnoses and therapy decisions. At the same time, 3D Histopath can be of relevance for pharmaceutical development. 3D Histopath is currently being developed by a medical doctor and researcher in human genetics at Charité and a lab technican. The team is supported by advisors in business matters, technology development, user research, and market access.
AKICHECK – Clinical Decision Support System to Identify Acute Kidney Injury
Prof. Prof. h.c. Dr. med. Markus van der Giet
Charité – Universitätsmedizin Berlin Medical Department, Division of Nephrology and Internal Intensive Care Medicine Campus Benjamin Franklin
1.7 million deaths per year are caused by Acute Kidney Injury (AKI) globally. AKI is a frequent clinical event occurring in up to 20% of all hospital patients. It is characterized by a rapid deterioration of renal function to varying degrees, and associated with an up to 15-fold increased risk of mortality. In addition, patients with AKI have a significantly higher risk of developing or exacerbating a chronic kidney disease. The standard of care for AKI detection in clinical routine is based on a biomarker measurement and takes 48 to 72 hours to yield results, thus leading to delayed therapeutic intervention. As of today, an early detection tool for AKI is not available. AKICHECK aims to close this gap of early detection with an easy-to-use tool for rapid and precise kidney function measurement. Translating scientific expertise in kidney function measurement to clinical routine, AKICHECK utilizes a proprietary database, a protocol for contrast agent measurement, and a software to diagnose AKI within the first two to seven hours – reducing the time needed over tenfold. AKICHECK is easy to integrate into today’s clinical workflow everywhere and, given low component costs, promises a step-changing improvement in both patient outcomes and healthcare system performance. Team AKICHECK unites deep expertise in clinical medicine with focus on nephrology, biomedical and laboratory expertise, biostatistics and machine learning. The team is working with advisors in the areas of software development, regulatory affairs, and market access.
ARCAS – AI for Life Sciences Best Treatment Possible for Every Cancer Patient
Dr. Altuna Akalin
Head of Bioinformatics and Omics Data Science Platform Berlin Institute for Medical Systems Biology (BIMSB), Max Delbrück Center (MDC) and Team
Cancer – a disease of the genome – is the second leading cause of death globally and is responsible for an estimated 9.6 million deaths in 2018. To make cancer treatments more effective it needs to be personalized from diagnosis to treatment. In today’s clinical practice, however, the information from the genome is either not used or is used inefficiently.
Team Arcas is building an AI-based diagnostic decision support system for cancer. This system makes sense of complex genomic information to support personalized, precise diagnosis and therapy recommendations. At the core, the system analyzes complex genomic information: every cancer biopsy is sequenced not only for mutation detection, but also for large-scale alterations, gene expression, and epigenetic changes. Arcas is using a multi-level deep learning approach to integrate clinical, genomic, and pharmacological data. With this system, Arcas can predict patient cancer subtypes, survival outcomes, and personalized drug response, more precisely. Arcas has shown promising results for colon, breast, and lung cancer. With more data available, the system can be used for many more cancerous diseases. The Arcas system also serves pharmaceutical R&D purposes, e.g., by identifying biomarkers to support the stratification of clinical trial participants, or by helping to interpret the molecular differences between responders and nonresponders to a particular pharmaceutical product. Furthermore, Arcas aims to improve cancer diagnostics and clinical therapy decisions. Team Arcas consists of international experts in the field of bioinformatics, omics data science, and medicine from the Institute for Medical Systems Biology at Max Delbrück Center for Molecular Medicine in the Helmholtz Association in Berlin. The team is supported by an advisor in life sciences and new venture development.
Open.IU – A Diagnosis and Therapy Solution for Adolescents with Internet Gaming Disorder
Dr. med. Olga Geisel and Team
Department of Child and Adolescent Psychiatry, Psychosomatic Medicine and Psychotherapy, Charité – Universitätsmedizin Berlin
In our digitalized world, a rising number of adolescents is being affected by internet gaming disorder (IGD). The WHO defines IDG as the inability to stop playing even though it interferes with other areas of a person’s life, such as family relationships, school, work, and sleep. These problems would typically persist for at least one year. IGD is a distressing medical condition, which leads to daily life dysfunction, is associated with psychological and psychiatric issues, and thus needs qualified care.
Team Open.IU has developed an online solution to provide diagnosis and treatment of IGD and co-occurring psychiatric conditions to parties involved. Open.IU consists of two components: an online diagnostic test, and an online counseling and therapy tool. The diagnostic test is based on a scientifically proven, standardized test for targeted screening, an in-depth evaluation for IGD and potential co-morbidities. The counseling and therapy tool provides access to licensed therapists and to modules based on cognitive behavioral therapy. With Open.IU, seeking professional guidance and medical attention for internet gaming disorder will be accessible to everyone instantly and without waiting times. This solution is a low-threshold, easy-to-use service, and makes mental health care accessible for everyone at any time. In addition to patients, Open.IU also provides information and recommendations to caregivers, such as families of affected adolescents. The Open.IU solution can be extended in the future to also cover other mental health conditions such as ADHD, alcohol addiction, anxiety, or depression.
Open.IU consists of an interdisciplinary team of experts in psychiatric and psychosomatic diseases, diagnostics and therapy for children and adolescents from Charité – Universitätsmedizin Berlin and School of Medicine at Hofstra/ Northwell in New York. The team is supported by advisors for computer game development, user research as well as reimbursement.
PREFREE – For Reducing Uncertainty in Pregnancy – A Decision Support Tool and Home Monitoring Solution
Prof. Dr. med. Stefan Verlohren
Charité – Universitätsmedizin Berlin – Department of Obstetrics Maternal-Fetal Medicine
Maternal mortality is unacceptably high. As a leading cause of maternal mortality, preeclampsia and related hypertensive disorders of pregnancy claim the lives of nearly 76,000 women and 500,000 babies worldwide every year. This lifethreatening complication of pregnancy can be prevented by better diagnosis, treatment, and monitoring.
Prefree is an AI-based decision support tool for physicians to identify pregnant women at risk for pregnancy complications, especially preeclampsia. This patient-centered solution was developed based on a clinical outcome database and interviews with patients and physicians. The solution aims to support physicians to identify the individual risk for preeclampsia, to decide whether to hospitalize patients in need and to allow patients with low risk to return to their homes. The decision support solution will be complemented by a remote monitoring system that enables women returning home to closely monitor their signs and symptoms in a close contact with their physicians.
Prefree intends to reduce the risk of false diagnosis, to avoid unnecessary hospitalization, and to reduce healthcare costs by patient-centered remote care within the support system and convenience of their homes. In the future, the system can add new technologies for additional benefits, be scaled to other regions with even higher prevalence of preeclampsia, and be extended to other pregnancy complications. Team Prefree consists of a team of medical doctors of the Department of Obstetrics from Charité – Universitätsmedizin Berlin and a PhD student with machine learning and software development expertise. The team is supported by advisors in machine learning, patient-centric care, business matters and reimbursement.
siloa – Solution for Digital Early Detection of Alzheimer’s Disease
Herlind Megges (M.Sc.), Silka Dawn Freiesleben (M.Sc.)
Charité – Universitätsmedizin Berlin, Department of Psychiatry – Campus Benjamin Franklin, Geriatric Medicine - Memory Clinic
Worldwide, at least 50 million people are believed to be living with dementia; a number projected to reach 82 million in 2030 and 152 million in 2050. Dementia is a syndrome associated with deterioration of memory, thinking, behavior, and the ability to perform everyday activities. Dementia has a substantial physical, psychological, social, and economic impact, not only on people with dementia, but also on their caregivers, families, and society at large. Alzheimer’s disease is the most common form of dementia. While no cure for Alzheimer’s is known today, it is being feverishly worked on. Early detection and lifestyle interventions are believed to improve quality of life.
Team siloa is working on a digital test for the early detection of Alzheimer’s disease to intervene in the disease progression. For the test, the team is developing a digital biomarker, combining software-based tests that engage brain areas known to be affected in very early stages of Alzheimer’s disease. The test will be initiated by a physician and then conducted by the patients in the comfort of their homes for 15 minutes per day over the span of a month. An Alzheimer’s probability score will then be transferred directly to the physician to maximize certainty for their patients and their caregivers. Siloa wants to enable a future where early detection of Alzheimer’s disease facilitates the diagnostic process and ensures that everybody receives the care they need as early as possible.
Team siloa consists of clinicians and researchers in geriatric medicine at the Memory Clinic at Charité. The team is working with experts in user research, software development, and reimbursement.
SUMUS – A Trustable Physiotherapy Guide for Patients Affected by Muscle Diseases
Univ. Prof. Dr. med. Simone Spuler and team
Charité – Universitätsmedizin Berlin and Max Delbrück Center for Molecular Medicine Clinic for Muscle Disorders
Muscular Dystrophy (MD) is a set of genetic conditions that gradually causes the weakening and breaking down of skeletal muscles, leading to an increasing level of disability. This progressive condition is often at first affecting a particular group of muscles and then deteriorates them over time. Some types of MD eventually affect heart muscles or breathing-related muscles, at which point the condition becomes life-threatening. MD is a rare disorder but also one of the most frequent genetic conditions affecting roughly 1 in 3,500 individuals worldwide. As of today, there is no cure for MD, but treatment including physiotherapy can help to manage the progression of symptoms.
Physiotherapy for patients with MD is fundamental, albeit needs to be highly personalized to ensure a sufficient level of muscle stimulation while not causing an overstimulation that could cause further damage of the muscles. In Germany and Europe, only a few physiotherapists are trained to provide this expert service. Worse, there is a significant barrier to accessing these professionals given the mobility constraints of many patients. In sum, patients affected by muscle dystrophy need personalized and engaging physiotherapy to maximize their quality of life and potentially decelerate the progressive condition. Team SUMUS is developing a virtual physiotherapist tool to engage and correctly guide muscle patients to a well-balanced life with the right amount of training. The individualized training syllabus is devised by a physician and a physiotherapist in close coordination with the patient. Part of the solution is the SUMUS Smartwatch application that tracks any active movement of the patient’s arms (initial prototype) in daily life and then advises the patient whether to train, which exercises to use, and to what extent. The physiotherapist and the physician can monitor progress continually via captured longitudinal data and adjust individual training plans with patient input as needed. This mutual feedback feeds into a self-learning algorithm to ensure continuous optimization of the patient’s fatigue monitoring and training. In the future, SUMUS could be extended by further technologies to also cover additional body parts, e.g., sensor shirts, and additional digital physiotherapy applications. With an increase in use of the SUMUS application, real-world evidence will be generated to develop, validate, and monitor new therapies and hopefully lead to a cure for MD. SUMUS combines interdisciplinary expertise in neurology, muscle dystrophy diagnosis and therapy research at Charité and MDC, physiotherapy, computeraided medical robotics, and game-based learning. The team is working with advisors in user experience, hardware and software development and reimbursement.
AIgnostics - Computational Pathology
Prof. Dr. Frederick Klauschen
Institute for Pathology CCM - Clinical Pathology, WG System Pathology, Charité – Universitätsmedizin Berlin
„The DHA enabled us to turn our research results examining AI-based cancer diagnostics into practical applications. Not only did we have the chance to develop a software that is currently tested live, but we were also exposed to the business, regulatory, and legal aspects of our project. Learning these challenging skills put us in the position to pursue a stringent development path towards launching a spin-off company in the near future. I highly recommend the DHA to fellow researchers and clinicians looking to generate digital health innovations for better patient health outcomes."
Microscopic evaluation of tissue samples performed by pathologists is the basis of diagnosing cancer as well as many degenerative, infectious and inflammatory diseases. Given our aging population, the increase in numbers and complexity of cancer cases worldwide, a global shortage of pathologists is imminent. As a result, the capacity of healthcare systems to perform histological diagnostics and determine the right therapy is under increasing pressure.
To address this challenge, team AIgnostics has developed an AI-based image analysis system to assist pathologists in standardized and quantitative automated tissue diagnostics. This solution promises to be more accurate and faster than today's manual gold standard. The system can also be applied to drug development, where pharmaceutical companies depend on accurate assessments of histological tissue samples in preclinical animal studies or in clinical trials to stratify patients in drug efficacy and toxicity analyses.
Team AIgnostics consists of experts in the fields of diagnostic and computational pathology from Charité - Universitätsmedizin Berlin, in cooperation with Prof. Klaus-Robert Müller, a globally renowned expert in machine learning from Technical University Berlin.
BodyTime - A New Diagnostic Assay to Assess the Internal Clock
Prof. Dr. Achim Kramer
Institute for Medical Immunology - Chronobiology, Charité – Universitätsmedizin Berlin
"DHA helped us to focus on issues scientists usually do not consider as too important - but they are important to move a technology towards market application. DHA was great in training and coaching us in a limited time frame with a high degree of knowledge and flexibility."
The biological clock is essential for health. The fast-paced and globalized lifestyles in modern society increasingly lead to the misalignment or disruption of our biological clock, which is associated with numerous common diseases such as sleep disorders, psychological disorders, metabolic syndromes, rheumatic disorders, cardiovascular diseases, and cancer.
In the emerging field of chronomedicine, team BodyTime addresses this medical need with a blood test (ChronoMarker) to determine the individual’s biological clock by profiling genetic biomarkers and applying machine-learning algorithms. While as accurate as currently established tests, this solution promises to be less complex, faster and more cost-effective. As the first application, the team is targeting sleep disorders in collaboration with sleep labs as a stand-alone tool or companion diagnostic for pharmaceutical or behavioral therapy options. Other opportunity areas include stratification of clinical trial cohorts, application to other disease types, and solutions directly empowering affected individuals.
Team BodyTime consists of experts in the field of chronobiology, medicine, and data-analysis / software development from Charité – Universitätsmedizin Berlin, supported by an expert in life sciences and new venture development.
Press and Media Coverage
Cardio Prime - Diagnosis and Therapy Planning Platform for Patients with Cardiovascular Diseases
Prof. Dr. Titus Kühne, Kay Brosien
Institute for Imaging Science and Computational Modelling in Cardiovascular Medicine, Charité – Universitätsmedizin Berlin
"The DHA provided us with an outstanding support system as we embarked on our journey to a solid transition from research to market with expertise, endurance, and connections." -Kay Brosien
The care path for cardiovascular patients, ranging from symptom detection to diagnosis to therapy and disease management, is fragmented. Providing each patient with the right type of care at the right place of care is a key challenge for each healthcare system. In a fragmented cardiovascular care path, quality and efficiency of care suffer.
Team Cardio Prime has developed an innovative Digital Health platform for diagnosis and therapy planning to inform and improve the care path for patients with cardiovascular diseases. The solution enables physicians working in cardiovascular and other specialties at hospitals, in specialist practices and in general practices to diagnose cardiovascular conditions earlier and make even more informed care path decisions. The first application is a stress test of cardiac and heart valve function without pharmaceutically or physical activity-induced stress. Other opportunity areas are complementary cardiovascular analyses and decision-support systems for cardiovascular diagnosis and therapy planning.
Team Cardio Prime consists of an interdisciplinary team of experts in cardiovascular disease diagnostics and treatments, physicists, engineers and software developers from Charité – Universitätsmedizin Berlin.
dentalXr.ai - Deep Learning for Dental Image Diagnostics
PD Dr. Falk Schwendicke, Dr. rer. nat. Joachim Krois
Dept. of Restorative and Preventive Dentistry, Charité – Universitätsmedizin Berlin
"The DHA wa a great opportunity for us. It allowed us to not only develop a strong team using the flexible financial support provided, but also to grow into a start-up, building the mindset and the network. We would not have reached our current status without this support. We can highly recommend the program!"
Dental diseases are among the most prevalent of mankind, burdening billions of individuals with pain, impaired chewing movements, impaired speech and aesthetics. To manage these diseases, early detection and regular monitoring with supportive therapy is needed. While visual-tactile detection has low sensitivity for early disease stages, the analysis of dental images is challenging due to limited accuracy of individual examiners, low consistency between examiners, and required time.
Team DentalXr.AI is developing an artificial intelligence (AI)-based decision-support system for dental images, intended to help dentists to systematically and comprehensively assess X-rays, document these assessments, and form evidence-based decisions. The envisioned solution enables faster and more accurate assessments of dental X-rays. This will result in reduced diagnostic uncertainty and better treatment decisions in less time. DentalXr. AI reduces assessment and treatment costs for patients and the healthcare system and makes dentists’ lives easier. Future application areas include incorporating additional image types and predictive capabilities.
Team DentalXr.AI consists of senior clinicians, machine learning experts and software developers from Charité – Universitätsmedizin Berlin.
LingPed - An Innovative Monitoring Platform for Post-Surgical Rehabilitation
Dr. med. Serafeim Tsitsilonis, Nevda Kaya
Center for Musculoskeletal Surgery, Charité – Universitätsmedizin Berlin
Today, the rehabilitation of patients after surgery of the lower extremity is mostly a black-box to the surgeon. The responsibility to control the loading of the operated extremity and to move in a safe manner rests with the patient. This can be overwhelming and complex for patients. Failure to adhere to the postoperative protocol leads to an increase of complications, re-operations, socioeconomic costs and can negatively affect the quality of patients’ lives.
Team LingPed is closing this gap in post-surgical rehabilitation with a monitoring system for patients after orthopedic surgery of the lower extremities. The system consists of an insole for use e.g. in postoperative shoes (orthoses) for data collection, and an app for patients as feedback mechanism to monitor and, if needed, adjust their behavior during recovery. The solution will also assist the surgeon in planning an individualized rehabilitation protocol. The system intends to reduce the risk of re-surgery, shorten and optimize the individual rehabilitation process, and reduce healthcare system costs. Future application areas include pattern recognition of gait abnormalities, telemedicine, prognosis and prevention of foot deformities for children and teenagers for optimal foot development.
Team LingPed consists of orthopedic trauma surgeons of the Center for Musculoskeletal Surgery from Charité – Universitätsmedizin Berlin and a medical student with hardware and development software expertise.
mTOMADY - A Transaction Platform for Accessible and Affordable Healthcare
Dr. med. Julius Emmrich, Dr. med. Samuel Knauss
Dept. of Neurology with Experimental Neurology, Charité – Universitätsmedizin Berlin
"The flexibility and willingness of the DHA to adapt to the needs and requirements of each individual team is the greatest strength of this program. It is not a one-size-fits-all approach. Instead, each team receives the support required to take the next step towards application."
More than 1 billion people in low- and middle-income countries lack access to basic healthcare. The majority of affected people do not have access to savings mechanisms and are at risk for unexpected expenses and even medical impoverishment. Recognizing this and related needs, promoting good health and well-being and alleviating poverty are UN Sustainable Development Goals.
Leveraging the global mobile phone infrastructure, team mTOMADY has developed a digital health wallet – a mobile transaction platform for healthcare savings. Healthcare sponsors, e.g., international aid organizations, governments, employers and communities, can contribute to individuals’ healthcare accounts, which in turn can only be used at accredited clinics within defined reimbursement ranges. This solution promises reduced “leakage” of aid funding by international aid organizations, quality improvement and cost-control for governments and healthcare providers, and – most importantly – increased access to affordable, quality healthcare for patients. mTOMADY is currently live with a pilot in Antananarivo in cooperation with the Government of Madagascar.
Team mTOMADY consists of two neurologists from Charité – Universitätsmedizin Berlin and a team of mobile technology, software development, and public health experts.
X-Cardiac: Artificial Intelligence-based Prediction of Complications in Intensive Care Units
Dr. med. Alexander Meyer
German Heart Center Berlin
"Networking and the exchange of opinions, targeted coaching, and sound mentoring: The Digital Health Accelerator Program from the Berlin Health Innovations is a genuine 'worry-free package' for successfully translating your ideas to patient solutions!"
Alexander Meyer is a computer scientist and senior resident in cardiovascular surgery. He is head of the research group “Data science in cardiovascular medicine” at the Department of Cardiothoracic and Vascular Surgery at the German Heart Center Berlin (Deutsches Herzzentrum Berlin). His main research interest is to improve patient safety with the application of state-of-the-art data science methodology in the medical domain.
PREDICTioN 2020 - Simulation-based AI for the Prediction of Stroke
Dr. Dietmar Frey
Dept. of Neurosurgery, Charité – Universitätsmedizin Berlin
"DHA supported us greatly in meeting next steps and milestones. By setting a common goal - the demo day - the whole team focused its effort on our shared mission: predicting stroke risk with artificial intelligence to save lives! This included aligning the development of the product with the user, analyzing the market, developing a comprehensive business strategy, and pitch training. In particular, the extensive and high-quality network enabled us to get feedback from different perspectives. All in all, the DHA supported us in taking our idea to the next level."
Team PREDICTioN 2020 develops a simulation program for stroke prognosis with the aim of enabling optimal treatment strategies for stroke prevention.
Each year more than 1.2 million people suffer from stroke in Europe alone. 15% of which have a re-stroke within a year. Prevention strategies and clinical treatment of stroke are based on generalized guidelines and empirical data alone. No personalized diagnostics and treatment strategies are available. Digital opportunities are not exploited sufficiently. With this AI solution Team PREDICTioN 2020 aims at saving lives of stroke patients by personalizing stroke prevention and treatment utilizing high-precision imaging and unique Machine Learning approaches.
The innovative simulation model enables predictive diagnostics in stroke treatment using digital medical imaging. Patient-specific MRI and CT images are processed with state-of-the art processing-tools and convolutional neural nets. This information is fed to the team's unique simulation that allows the estimation of cerebral perfusion changes under different conditions.
The simulation results together with routine clinical data allow predictive modelling for stroke risk and outcome using innovative machine learning techniques. The approach will lead to personalized simulation of therapy options in stroke enabling patient-centered treatment and personalized risk assessment.
Team PREDICTion 2020 consists of neurosurgeon Dr. Dietmar Frey from Charité - Universitätsmedizin Berlin and a team of software developers, machine learning experts and scientific coordinators.