News and Events

CAIM in the Media


Sign up here if you would like to receive CAIM Updates. // CAIM Updates Archive

Interpretable ML system for colorectal cancer diagnosis (05.03.2024)

With engineers and clinicians from Portugal, the Digital Pathology group of the Institute for Tissue Medicine and Pathology, University of Bern, has developed a scalable deep learning system to diagnose colorectal cancer from pathology images. The system learns from weak labels, leverages a small subset of fully annotated samples, and the prototype includes explainable predictions and active learning. Developed and tested with one of the largest colorectal samples datasets, it showed high to very high accuracy.

MLP mixers in lung image segmentation (07.02.2024)

The AI in Health and Nutrition lab at the ARTORG Center conducted performance tests on a computer-aided diagnosis (CAD) system for lung fibrosis, employing multi-layer-perceptron-mixers (MLP-mixers) for segmenting lung and airway anatomy. Additionally, the system identifies idiopathic lung disease patterns in a second step. Currently, MLP-mixers segmentation demonstrates performance comparable to nnU-Net on chest CT images.

Pfizer prize for AI-based prediction for recovery from coma (25.01.2024)

Florence Aellen, Athina Tzovara, and their colleagues from Cognitive Computational Neuroscience, University of Bern and Inselspital, have been awarded one of the coveted Pfizer Research Prizes in Neuroscience and Neurology. They challenged the traditional approach for predicting coma outcome which leaves a percentage of patients without clear prognosis, and used deep learning algorithms applied to coma patients’ EEG responses to sound stimuli to assess the integrity of neural functions in coma and chances of recovery.

Active learning for multi-label xray classification (11.01.2024)

In medical image data sets with limited labeled data, active learning approaches can help reduce annotation costs and boost performance of computer-aided diagnosis systems. Based on its previous findings employing interpretability information as inductive bias and to select informative samples in active learning, the Medical Image Analysis lab has now developed a new graph-based transformer and data augmentation active learning framework for multi-label xray classification.

Saliency feature learning for MRI (26.10.2023)

Introducing saliency feature learning to multi-sequence MRI could improve the segmentation quality of deep learning systems. This is the finding of a recent study of the Medical Image Analysis lab in collaboration with the Support Center for Advanced Neuroimaging (SCAN) and the Department of Radiation Oncology of the Inselspital, Bern University Hospital.

Personalized lab data readings across hospitals (18.10.2023)

Researchers from the BioRef consortium of the University Hospitals of Bern (Inselspital) and Lausanne (CHUV), the University Children's Hospital Zurich and Swiss Paraplegic Research have established a multicentric IT framework together with Tune Insight and SPHN to obtain patient group-specific reference intervals from routine laboratory data. This infrastructure, BioRef TI4Health, is the first of its kind featuring a federated network where patient data remains at each participating hospital, yet allowing a joint evaluation in real-time, reproducible, and secured by homomorphic encryption.

Monitoring nutrition with a single image (04.10.2023)

In a refinement of its earlier system for automated nutrient assessment via smartphone, the AI in Health and Nutrition laboratory in collaboration with the School of Health Professions, Bern University of Applied Sciences, has developed an easy-to-use single-image-input pipeline to output nutrient composition within the goFOODTM application. In a retrospective real-world study, the new system reduced user burden and showed promising results.

Multiclass AI model rates Covid severity (29.09.2023)

To improve the initial patient assessment and disease characterization from lung CT scans, the Medical Image Analysis lab of the ARTORG Center, Uni Bern, and the Department of Radiology of the Inselspital, Bern University Hospital have developed a 2-stage multiclass lung lesion model trained to classify disease severity based on the WHO Clinical Progression Scale. The models performance exceeded that of a single-class model as well as the radiologists’ assessment.

AI-assisted brain tumor measurements (19.09.2023)

Automated tumor segmentation for glioblastoma is promising, yet little is known about longitudinal accuracy of automated measurements to assess treatment response. The Medical Image Analysis lab at the ARTORG Center for Biomedical Engineering Research at the University of Bern compared assessment consistency by two AI-based tumor segmentation tools against expert ratings over time, identifying a need for further improvement.

Machine learning limited in subclassification of inflammatory bowel disease (13.09.2023)

The Inselspital and ARTORG Center, Uni Bern, have tested a supervised and unsupervised machine learning approach using antibodies in lab samples to distinguish Crohn’s disease (CD) and ulcerative colitis (UC) as subtypes of inflammatory bowel disease (IBD). The approach showed good sensitivity to correctly assign CD and UC, but could not aid in finer distinction of subtypes such as IBD-unclassified due to the lack of specificity of the analyzed markers.

Deep Learning Dose Prediction for Radiotherapy (11.09.2023)

Radiotherapy treatment requires accurate contouring of tumor volumes and organs at risk. In a collaboration between the ARTORG Center for Biomedical Engineering Research, Uni Bern, and the Department of Radiation Therapy, Inselspital, a deep learning architecture for automatic segmentation and 3D dose prediction for glioblastoma patients has shown good sensitivity and robustness.

CAIM Fellowships Call 2023 (20.07.2023)

The CAIM Research Fund is opening a call for CAIM Fellowships. The project funding to support early-career researchers with promising career prospects in translational AI in medicine research will support projects for up to 2 years with up to 100’000 CHF. Deadline for applications is 25 September 2023.

AI assisted epilepsy diagnosis (13.7.2023)

In a clinical validation, the Support Center for Advanced Neuroimaging (SCAN) of the Inselspital has compared the accuracy of combining an AI assistant developed at SCAN with a human expert reading of MRI imaging with the performance of a neuroradiologist reading to diagnose hippocampal sclerosis. The team found that the "human + machine" model improved the mean accuracy of the raters from 77.5% (MRI only) to 86.3% (MRI + AI assistant).

Statistical learning in lab data (20.06.2023)

Laboratory information systems are struggling to manage and analyze the amount of complex and entangled data generated in laboratory medicine today. Statistical learning, a generalized framework from machine learning and AI, is predestined for processing this “Big Data” and holds the potential to revolutionize the field. To ensure quality, transparency, and public acceptance, human experts should carefully validate AI-based systems, including patient-privacy protection.

Tumor budding T-cell graphs (05.06.2023)

Minimally invasive types of colorectal cancer (pT1) may only necessitate removing the polyps instead of requiring full colon resection to stop cancer progression. To assist pathologists, the Institute for Tissue Medicine and Pathology of the University of Bern has evaluated a Graph Neural Network based approach for risk stratification based on the interaction of tumor buds and T-cells as well as other histopathological risk factors with about 20% higher specifitiy than current guidelines.

AI in interstitial lung disease (30.05.2023)

Over 200 diagnoses exist for the umbrella term "interstitial lung diseases" (ILDs). These are currently determined by an expert board based on CT imaging, patient data, pulmonary function tests, and histology. AI methods could help aid in accurate ILD diagnosis and predict prognosis and progression in a holistic system, as a review paper by the ARTORG Center and Inselspital comparing current approaches shows.

AI to detect lymph node metastases in colorectal cancer (03.03.2023)

Screening of lymph node metastases in colorectal cancer (CRC) could be solved by an AI-assisted diagnostic solution. A group of pathologists and data scientists led by the Institute for Tissue Medicine and Pathology, University of Bern, has proposed a deep learning–based workflow for the evaluation of CRC lymph node metastases. The approach showed very good performance, providing the basis for a computer-assisted diagnostic tool for easy and efficient lymph node screening in CRC patients.

Measuring blood glucose with the smartwatch (23.02.2023)

Researchers at the Inselspital, Bern University Hospital, and the University of Bern developed a method that uses machine learning and smartwatch data to detect hypoglycemia. The approach was tested in a pilot study with 31 diabetes patients and could in the future be used to non invasively measure blood glucose levels as a valuable complement to existing methods for blood glucose monitoring.

Medienmitteilung Insel Gruppe, 23. Februar 2023 (PDF, 119KB)

Bridging deep learning and clinical radiotherapy (08.02.2023)

The ARTORG Center Medical Image Analysis group (MIA) and the Department of Radio-Oncology of the Inselspital have developed an open-source, deep learning framework-independent Python package feasible for processing DICOM RT Structure Sets, in collaboration with Varian a Siemens Healthineers Company. “PyRaDiSe” (Python (Py), radiotherapy (Ra), DICOM-based (Di), auto-segmentation (Se)) goes beyond current 2D reconstruction, potentially supporting acceptance by healthcare professionals.

Predicting heart attack from routine hospital data? (04.01.2023)

The Department of Clinical Chemistry and the Insel Data Science Center at the Inselspital have conducted a proof-of-concept study to infer the diagnosis of myocardial ischemia from predictive analytes extracted from routine hospital lab data. Authors could demonstrate that an assembly of algorithms was able to create unbiased guidelines for laboratory diagnostics by using computational evidence in today’s era of healthcare digitalization. This is an important first step towards evidence-based laboratory diagnostics and, therefore, improved patient care.

AI-assisted interpretation in cardiotocography (30.12.2022)

Artificial intelligence is gaining interest in the field of cardiotocography interpretation in obstetrics, since it promises to remove existing biases and improve the well-known issues of inter- and intra-observer variability. A review publication by the Department of Obstetrics and Feto-maternal Medicine, Bern University Hospital, with CSEM hints at potentials of AI to help detect complications in the unborn child earlier.

LUMIERE Dataset for Glioblastoma Released (21.12.2022)

The Medical Image Analysis group at the ARTORG Center of the University of Bern in collaboration with colleagues at Inselspital, Bern, has released the first publicly available single-center dataset of Glioblastoma Magnetic Resonance Imaging (MRI) with expert ratings. It comprises over 600 multi-sequence MRI acquisitions and expert readings of selected studies according to the response assessment in neuro-oncology (RANO) criteria, automated tumor segmentations, and a rich set of complementary information, such as advanced imaging biomarkers, patient demographics and pathology information.

Deep Learning detects neurodegeneration at the micrometer scale (16.12.2022)

A research group from Inselspital and the University of Bern has developed a deep learning tool that can measure neuro-degeneration reliably and significantly faster than established methods. A scientific paper now shows that the algorithm can be trained to read different types of MRI images used in everyday clinical practice. Researchers at the CSIRO's Australian eHealth Research Center have furthermore shown in an independent evaluation that the open-source method can measure subtle changes in the range of up to 0.01mm (reference 0.06mm).

Averting severe embolisms through artificial intelligence (1.12.2022)

Experts in laboratory medicine at Inselspital, Bern University Hospital, have developed a diagnostic prediction model in a long-term prospective multicenter study in collaboration with data scientists at the University of Bern, which provides treating physicians with an indicator to reliably diagnose HIT (heparin-induced thrombocytopenia) - a rare but life threatening complication for hospitalized patients after major surgery or severe inflammation. The diagnostic machine learning tool derived from this research in the field of laboratory medicine is unique in the world.

Media release Insel Gruppe, 1 December 2022 (PDF, 154KB) Medienmitteilung Insel Gruppe, 1. Dezember 2022 (PDF, 105KB)

Great Interest in the CAIM Research Symposium (28.11.2022)

The first Research Symposium of the Center for Artificial Intelligence in Medicine since its official opening in March 2021 was attended by more than 100 researchers from the Bernese ecosystem around the University and University Hospital. CAIM Director Raphael Sznitman was excited to witness the beginning of promising interdisciplinary scientific discussions in the wake of the two keynote speeches and during the networking breaks: "It is great to have such a vibrant community around AI research for healthcare here in Bern."

Amith Kamath and Charlotte Kern win CAIM Young Researcher Awards (25.11.2022)

The very translational research of PhD students Amith Kamath, ARTORG Center, University of Bern, and Charlotte Kern, Clinical Pharmacology and Toxicology at the Bern University Hospital, has been recognized at the CAIM Research Symposium via the new CAIM Young Researcher Awards. Awards were sponsored by the Bern Economic Development Agency with a total of CHF 1000. The research developed yet a bit further into a product (Amith Kamath) additionally received 16 hours worth of personalized entrepreneurial coaching by be-advanced.

Using interpretability to improve medical AI (24.11.2022)

The Medical Image Analysis lab at the ARTORG Center has just published a study where interpretability informed sample selection from a set of radiological images. The lab, which has focused on interpretability both as a means for more transparency in medical AI workings and as a tool to improve AI itself, reports that their novel sample selection approach based on graph analysis to identify informative samples in a multi-label setting improved model performance, learning rates, and robustness when compared to state-of-the-art Active Learning methods.

Less radiation with AI-enhanced PET (03.11.2022)

PET/CT can be used to clarify tumor diseases regarding their extension and therapy response, as well as to answer neurological and cardiological questions. The Bern University Hospital uses the world's most sensitive PET scanners with comparatively low radiation exposure. AI-supported image attenuation could further lower radiation and result in better imaging, as demonstrated by the Nuclear Medicine Department in Nature Communications.

Media release Insel Gruppe, 3 November 2022 (PDF, 155KB) Medienmitteilung Insel Gruppe, 3. November 2022 (PDF, 153KB)

Maintaining a Mediterranean diet with your phone (19.10.2022)

Researchers at the ARTORG Center in collaboration with Oviva and with the Epidemiology, Biostatistics and Prevention Institute (EBPI), University of Zurich, have demonstrated that an AI system on a smartphone can effectively track whether a person is adhering to a Mediterranean diet (MD). MD can decrease the risk of non-communicable disease and prevent overweight and obesity. The feasibility study of the system was just published in Nature Scientific Reports.

Bernese AI system receives FDA approval (30.09.2022)

An artificial intelligence system for brain tumor segmentation from multisequence MRI has received 510k clearance by the US American Food and Drug Administration on 29 September 2022. The system was developed by Prof. Dr. Mauricio Reyes and his team from the Medical Image Analysis lab, at the ARTORG Center and the Center of AI in Medicine, Univ. of Bern. Work on this important milestone by many talented PhD and Master students has started at the @MICCAI 2011 conference, receiving the 2016 MICCAI Young Scientist Publication Impact Award as well as the 2016 Ypsomed Innovation Award. It underlines the importance of Bernese research on a global scale.

AI: Friend and helper of medical professionals? (21.09.2022)

AI is profoundly changing the healthcare system. This process poses various challenges for the medical profession. With the brochure "Artificial intelligence in everyday medical practice", the Swiss Medical Association FMH offers the medical profession an overview of AI in the medical environment. Ten demands raise awareness for the essential aspects of AI in medicine.

Medienmitteilung FMH, 21. September 2022 (PDF, 104KB) Communique de presse, FMH, 21 septembre 2022 (PDF, 103KB)

Making the most out of lab data (14.09.2022)

Every large hospital produces a wealth of data each day, holding the key to new insights into diseases and improved patient care. For machine learning and Big Data approaches to be utilized on this data, lab data needs to be FAIR: findable, accessible, interoperable, and reusable, propose a team of researchers from the Department of Clinical Chemistry at the Inselspital.

Sensor-based early detection of age-related diseases from home (30.08.2022)

Sensors that record movement patterns could help detect health problems in the elderly, including old-age depression, risk of falls or cognitive impairment, at an early stage. In the future, this could help seniors to live a self-determined life at home for longer and relieve increasing pressure on the healthcare system.

Media release University of Bern, 30 August 2022 (PDF, 90KB) Medienmitteilung Universität Bern, 30 August 2022 (PDF, 116KB)

Hypo- and Hyperglycemia Prediction (08.08.2022)

Blood glucose decompensations of hospitalized patients poses a frequent and significant risk for patient outcomes and safety. A team of data scientists and doctors from the University of Bern and Bern University Hospital have generated a broadly applicable multiclass classification model for predicting decompensation events from patients’ electronic health records to indicate necessary adjustments in monitoring or therapy.

ARTORG Center to develop digital care assistant with QUMEA (01.07.2022)

Together with digital health startup QUMEA, the ARTORG Center for Biomedical Engineering Research, University of Bern and the Bern University of Applied Sciences (BFH) have started a project to enable hospital care teams and doctors to adapt and improve care via sensor-based technology. The project is funded by an Innosuisse grant of CHF 1.2 million under the impulse program Swiss Innovation Power.

AI for diabetes project MELISSA launched (09.06.2022)

The EU Research Project “MELISSA: Mobile Artificial Intelligence Solution for Diabetes Adapted Care” was launched today by a consortium of 12 partners, comprising the ARTORG Center Artificial Intelligence in Health and Nutrition (AIHN) lab, University of Bern. AIHN is the artificial intelligence expert in the project and initiated it in collaboration with Debiotech.

MELISSA press release, 9 June 2022 (PDF, 353KB)

Strange dreams help your brain learn better (18.05.2022)

A new study by researchers from the University of Bern, Switzerland suggests that dreams—especially those that simultaneously appear realistic, but, upon a closer look, bizarre—help our brain learn and extract generic concepts from previous experiences. The study, carried out within the Human Brain Project and published in eLife, offers a new theory on the significance of dreams using machine learning inspired methodology and brain simulation.

Predicting Urinary Infection with AI (05.05.2022)

In a special Issue on Urinary Stones and Infections of Diagnostics, the Inselspital, the University of Cologne and the University of Bern have published a recent AI-guided analysis of urine samples from emergency patients. The assessment of 3835 samples revealed that AI can safely rule out urinary infections in 96 percent of cases. However, for a high-confidence AI prediction of positive infections and infection types, using UFC parameters alone is not sufficient.

Cortical oscillations in neural networks (06.04.2022)

Dealing with uncertainty is a big challenge for our brain. Led by the University of Bern, researchers from Heidelberg University, the Jülich Research Centre and Graz University of Technology have identified a new computation role for cortical oscillations that support sampling-based computations in spiking neural networks.

Inti Zlobec elected Digital Pathology Professor (31.03.2022)

The University of Bern has elected Inti Zlobec as Professor of Digital Pathology as of April 2022. The position is one of several new professorships in the field of artificial intelligence and digitalization in medicine that have been created as part of a new research focus in this area.

AI enables personalized treatment of myocarditis (24.03.2022)

A research team from the University of Bern and Inselspital, University Hospital Bern is investigating and developing innovative approaches that will enable personalized diagnosis and treatment of myocarditis. Artificial intelligence will allow individual risk assessment and progression prognosis in the future. The project has received funding from the CAIM Research Project Fund for 2022/23.

Media release Insel Gruppe, 24 March 2022 (PDF, 133KB) Medienmitteilung Insel Gruppe, 24. März 2022 (PDF, 233KB)

CAIM supports Ukrainian scientists (10.03.2022)

The Center for AI in Medicine supports scientists from Ukraine in these difficult times. Following the measures and call from the Swiss National Science Foundation to help Ukrainian researchers as well as the University of Bern’s initiative to solidarize with Ukraine and its universities, CAIM calls its members and partners to reach out and proactively seek ways to help our colleagues.

Bayesian prevalence estimation in a nosocomial outbreak (09.03.2022)

Nosocomial pathogens in hospitals require strict screening and prevention. To better understand their dynamics and the impact of control measures, a team of infectiologists, data scientists and clinical chemists from the Inselspital applied a Bayesian hierarchical model to a nosocomial outbreak. Authors conclude that such models provide a more flexible platform to study transmission dynamics.

CAIM funds five translational projects (10.02.2022)

Twenty teams submitted their projects to the first CAIM Research Projects Fund call. After a peer-reviewed assessment by external reviewers and a pitching round, the CAIM Management and Steering Committee have selected five projects to be funded with up to CHF 100,000 each over the coming 24 months.

CNN quantifies ureteral stent encrustations (08.02.2022)

Location and extent of encrustation in ureteral stents for different diseases may be informative for patient management and the development of newer stent generations. The Urogenital Engineering Lab at ARTORG has investigated stent encrustation patterns from stone and kidney transplant patients in collaboration with the Inselspital and the Cantonal Hospital Olten, quantifying them via a Convolutional Neural Network (CNN) model.

AI to prevent malnutrition in hospitals (17.01.2022)

During hospitalization, especially older patients are at risk of eating too little, leading to loss of muscle mass and higher risks of infection and mortality. The AI in Health and Nutrition lab, ARTORG Center, has introduced an AI-powered system to automatically assess each patient’s energy and macronutrient intake by comparing pictures taken before and after each meal. In collaboration with the Geriatrische Klinik St. Gallen and the Inselspital, the team has assessed the systems’ performance.

Bern Data Science Initiative launched (17.12.2021)

The Bern Data Science Initiative, BeDSI, is an initiative of data driven research groups and research centers of the University of Bern. It is designed to weave a flexible network lead by scientists into the university’s traditional structures. BeDSI was launched on December 6, 2021. In an interview three initiators, Christiane Tretter, Raphael Sznitman and Tobias Hodel, explain why this network is so important for the future.

PeriVision wins EIT Wild Card 2021 (15.12.2021)

PeriVision, a MedTech spin-off from the ARTORG Center (AI in Medical Imaging lab) has won a prestigious European Wild Card in the 2021 challenge. The Bernese startup will be receiving an investment of EUR 1.5M along with specific coaching, which the founders will use to finalize their first product and prepare its EU and US market launch. PeriVision aims to reimagine glaucoma patient monitoring by making visual field testing much more patient-friendly and cost-efficient.

University of Bern at the NeurIPS 2021 (07.12.2021)

Paul Haider from the Computational Neuroscience Group of the Physiology Department, University of Bern, will hold an oral presentation at the Thirty-fifth Conference on Neural Information Processing Systems (NeurIPS 2021) entitled "Latent Equilibrium: A unified learning theory for arbitrarily fast computation with arbitrarily slow neurons”.

CAIM launches Expert Directory (01.12.2021)

To map the wealth of know-how in developing, validating, and translating AI technology into healthcare inside the Bern Medical Hub, CAIM is launching an Expert Directory. A first version of the Directory goes online today with a non-exhaustive list of clinical and data science experts from the University of Bern, Inselspital, UPD and sitem-insel. Registration is rolling and can be done at

When algorithms get creative (10.11.2021)

Uncovering the mechanisms of learning via synaptic plasticity is a critical step towards understanding how our brains function and building truly intelligent, adaptive machines. Researchers from the University of Bern propose a new approach in which algorithms mimic biological evolution and learn efficiently through creative evolution.

Medienmitteilung Universität Bern, 10 November 2021 (PDF, 38KB) Communique de presse, Université de Berne, 10 novembrre 2021 (PDF, 128KB)

Neuromorphic deep learning with first spike times (20.10.2021)

Researchers at Heidelberg University and University of Bern have recently devised a technique to achieve fast and energy-efficient computing using spiking neuromorphic substrates. Just as energy consumption and reaction times are critical to a biological agent operating under environmental pressure, engineered systems need to achieve the desired results with as few and as early spikes as possible, thus optimizing their short time-to-solution and low energy-to-solution characteristics.

Link to the study
News article TechXplore, 5 October 2021
Media release University of Heidelberg, 29 October 2021

Illustration of the on-chip classification process with the Yin-Yang dataset. Each symbol represents the spike time delay for various classifying neurons (J. Göltz & L. Kriener et. al.)

Pulmonary fibrosis: reliable prognosis thanks to AI (14.10.2021)

A research team from the Universities of Zurich, Oslo and Bern and their respective University Hospitals has published results on AI-supported image analysis of pulmonary fibrosis, which occurs in rare systemic sclerosis. By applying methods of radiomics analysis, the researchers, led by principal investigator Prof. Britta Maurer, have produced surprisingly clear risk profiles that offer a promising basis for future personalized patient management.

Medienmitteilung Insel Gruppe, 14. Oktober 2021 (PDF, 191KB) Media release Insel Gruppe,14 October 2021 (PDF, 391KB)

Advancing AI image interpretation for clinical use (01.10.2021)

AI is increasingly used in image interpretation for diagnosis and treatment planning. A research team from Inselspital, Bern University Hospital and the University of Bern demonstrated in a study that the current methods of qualifying AI for brain segmentation could be enhanced. Deviations of currently used parameters do not correlate with clinically relevant changes of the radiation dose distribution. For wide implementation of AI based software with real added value in treatment quality a stronger focus on clinically relevant outcomes is needed.

Deep learning predicts subtypes in colorectal cancer (20.09.2021)

The research team of Prof. Zlobec, Institute of Pathology, University of Bern, has proposed a deep learning method for quantifying extracellular mucin-to-tumor area in colorectal cancer (CRC). This histomorphology feature is an important indication to predict consensus molecular subtypes and clinical outcome. The study contributes to their final research goal: improve CRC patient stratification, prognosis and treatment prediction using deep learning and multi-omics data.

Opening of NeuroTec research facility (15.09.2021)

The NeuroTec research facility was opened by Insel Gruppe, together with the University of Bern and sitem-insel, on the Insel Campus in Bern on Tuesday, 14 September 2021. NeuroTec pools the clinical knowledge of Inselspital, University Hospital Bern, and the expertise in medical technology of the University of Bern, while further expanding Bern’s leading position in neurology pioneered by Prof. Dr. Claudio Bassetti and his team.

Medienmitteilung Insel Gruppe/NeuroTec, 15. September 2021 (PDF, 312KB) Media release Insel Gruppe/NeuroTec, 15 September 2021 (PDF, 393KB)

Bayesian Brain for Tinnitus (10.09.2021)

The Hearing Research Lab of the Inselspital, Bern University Hospital, and the ARTORG Center has developed a generative computational model for tinnitus, based on the Bayesian brain concept. The model is able to explain several perceptual tinnitus phenomena that are to date poorly understood. It can be applied for future research and treatment approaches by linking experimental observations with theoretical hypotheses.

User Preferences of Nutrition Apps (03.09.2021)

The Artificial Intelligence for Health and Nutrition lab at the ARTORG Center has conducted a study of adult smartphone app users for health and diet monitoring. In collaboration with the Inselspital and CentraleSupélec, the researchers sought to explore the perspectives of end users on the features, current use, and acceptance of nutrition and diet mHealth apps with a web-based survey. Goal was to provide user insights to assist researchers and developers in building innovative dietary assessments.

AI-assisted personalized radiotherapy (05.07.2021)

With new premises, the Department for Radio-Oncology at Inselspital is becoming a state-of-the-art center for the radio-oncological treatment of cancer patients. Three latest-generation devices are available for precise radiation therapy. With Varian's "Ethos", Inselspital is the first center in Switzerland to implement AI-assisted adaptive radiotherapy for patient-centered, personalized medicine. This opens up new opportunities for collaboration and research avenues in AI technologies in the framework of CAIM.

AI improves speech understanding of hearing aid users (30.06.2021)

In noisy environments, it is difficult for hearing aid or hearing implant users to understand their conversational partner because current audio processors still have difficulty focusing on specific sound sources. In a feasibility study, researchers from the Hearing Research Laboratory, at the University of Bern ARTORG Center and the Inselspital ORL Department are now suggesting that artificial intelligence could solve this problem.

Medienmitteilung Insel Gruppe, 30. Juni 2021 (PDF, 259KB) Media release Insel Gruppe, 30 June 2021 (PDF, 484KB)

AI could soon tell you, how often to see the eye doctor (08.06.21)

Three of the most common chronic eye conditions require regular medical check-ups and injections into the eye by ophthalmology specialists to keep looming blindness at bar. A study by the University of Bern and the Inselspital in collaboration with an AI in eye care startup now demonstrates that patients’ individual ideal frequency for these visits can quite accurately be predicted by machine learning – yielding a threefold benefit.

Link to the study

Study setup (© Artificial Intelligence in Medical Imaging Lab, ARTORG Center)

Medienmitteilung Insel Gruppe, 8. Juni 2021 (PDF, 284KB) Media release, Insel Gruppe, 8 June 2021 (PDF, 186KB)

Bern Data Science Day established (01.06.2021)

Data science research is on the rise at the University of Bern. For an initial stocktaking of ongoing projects in various faculties, the first Data Science Day was recently held. Following the great success of the pilot edition, the event is now to be held annually.

Space technology in the operating theater (12.05.2021)

The instruments were developed to search for signs of life on other planets. But now they will also be used to distinguish healthy nerve cells from brain tumor cells. A conversation with astrophysicist Brice-Olivier Demory and Raphael Sznitman, an expert in machine learning and artificial intelligence, about their BrainPol project. Ori Schipper reports on a new collaboration project between the Center for Space and Habitability and the ARTORG Center AI in Medical Imaging lab.

Center for Artificial Intelligence in Medicine opens (19.03.2021)

The Center for Artificial Intelligence in Medicine (CAIM) of the University of Bern and the Insel Gruppe with the partners sitem-insel and the Bern University Psychiatry Services UPD will be officially inaugurated today. The virtual opening event with 500 registered participants offers insights into controversial topics and current research projects on Artificial Intelligence in Medicine.

Medienmitteilung, Universität Bern, 19. März 2021 (PDF, 79KB) Media Release, University of Bern, 19 March 2021 (PDF, 95KB) Interview Raphael Sznitman, UniAktuell, 19 March 2021 (PDF, 2.2 MB)

Using AI to assess surgical performance (12.03.2021)

Lavanchy & Beldi, Insel Gruppe

A research team at Inselspital, Bern University Hospital, University of Bern and Caresyntax has succeeded in proving that artificial intelligence can reliably assess surgeons’ skills. A method involving a three-stage procedure has been presented that correctly designates good and mediocre performance with a high accuracy rate. This paves the way for further steps towards AI-supported expert systems.

Medienmitteilung, Insel Gruppe, 12. März 2021 (PDF, 124KB) Media release, Insel Gruppe, 12 March 2021 (PDF, 138KB)

Efficient treatment of stroke with AI and federated learning (09.03.2021)

A research team of Inselspital, University Hospital Bern, the University of Bern and the «Centre hospitalier universitaire vaudois» (CHUV) uses AI to improve efficiency of treatment after stroke. The project Advanced Stroke Analysis Platform (ASAP) applies federated learning that links the database of each hospital center without exchange of raw data. Innosuisse, the Swiss innovation agency, promotes the project.

ASAP federated learning setup (JPG, 46KB) Medienmitteilung Insel Gruppe, 9. März 2021 (PDF, 165KB) Media release Insel Gruppe, 9 March 2021 (PDF, 175KB)

Reliably predicting progression of Covid-19 (03.03.2021)

Inselspital, Bern University Hospital and the University of Bern are currently launching the world’s first multicenter, international study on AI-assisted prediction of severe progressions of Covid-19. The research uses artificial intelligence to evaluate extensive clinical, image-morphological and laboratory data. The study, which is funded by the Swiss National Science Foundation, aims to provide reliable predictions as to whether a specific case would lead to a severe progression of Covid-19.

Multi-omics approach (JPG, 149KB): Complex data, consisting of CT, X-ray data, clinical and laboratory data, serve the AI algorithm as a basis for the prognosis of the acute (7-day) and chronic course. (Alexander Poellinger, Inselspital).

News report Inselspital, 3 March 2021

Deep learning reads tissue microarrays of colorectal cancer (10.02.2021)

Tissue microarray (TMA) core images are a treasure trove for AI applications. But a common problem with TMAs is multiple sectioning of the TMA block, which can change the content of the core and requires re-labelling of each spot image. The Institute of Pathology of the University of Bern has investigated different ensemble methods for colorectal tissue classification using high-throughput TMAs. Their high-accuracy algorithm for colorectal tissue classification into “tumor“, “normal“ or “other” tissue showed very good predictive accuracy and is amenable to images from different institutions, core sizes and stain intensity.

A NEAT reduction of complex neuronal models accelerates brain research (27.01.2021)

Unlike their simple counterparts in AI applications, neurons in the brain use dendrites – intricate tree-like branches – to find relevant chunks of information. Now, neuroscientists from the University of Bern have discovered a new computational method to make complex dendrite models much simpler. This could enable AI applications to process information much like the brain does.

Medienmitteilung Universität Bern, 27. Januar 2021 (PDF, 65KB) Media Release University of Bern, 27 January 2021 (PDF, 122KB)

AI for quality assurance in radiation therapy planning: Project funded (18.01.2021)

The Medical Image Analysis group at the ARTORG Center, University of Bern, received funding support from the Swiss Cancer League to work on AI technologies for quality assurance in Radiation Therapy planning, in partnership with the Radiation Oncology Departments of the USZ and the Inselspital.

PI Mauricio Reyes comments: “We are thankful for this support as it has the potential to create a new generation of technologies dedicated to improve quality assurance aspects in radiation therapy planning.” The project ranked third out of seventy seven submitted projects.

Medical Imaging Analysis group, ARTORG Center

Figure Segmentation Dose Combination (MIA) (JPG, 493KB)

Thomas Sauter receives endowed professorship in emergency telemedicine (15.01.2021)

(photo: Pascal Triponez © University of Bern)

Thomas Sauter, Head of Education, eHealth and Emergency Telemedicine at Bern University Hospital's emergency center has been appointed by the University Executive Board for an endowed professorship in emergency telemedicine. The assistant professorship was established thanks to the support of the Touring Club Switzerland (TCS). It deals with "eHealth" in the field of emergency medicine and is one of the very few of its kind in the world.

Medienmitteilung, Universität Bern, 15. Januar 2021 (PDF, 90KB) Media Release, University of Bern, 15 January 2021 (PDF, 144KB) Communique de Presse, Université de Berne, 15 janvier 2021 (PDF, 127KB)

Sinergia project on trans-omics against colorectal cancer to start (06.01.2021)

The University of Bern Institute of Pathology has won a Swiss Science Foundation Sinergia grant to develop an integrative computational and clinical perspective on colorectal cancer, an important cause of cancer-related mortality in Switzerland. In collaboration with the ZHAW, the Institute of System Biology, ETH Zürich and IBM Research, the group will improve patient stratification, prognosis and treatment prediction via an AI-driven multi-modal classifier trained to predict molecular classification and identify other important clinical variables.

Smartphone verifies compliance with Mediterranean diet (29.12.2020)

The ARTORG Artificial Intelligence in Health and Nutrition lab, the Epidemiology, Biostatistics and Prevention Institute of the University of Zurich and Oviva SA have together developed a smartphone application to automatically estimate whether a person adheres to a health-promoting Mediterranean diet. First results of the novel system are promising.

AI X-ray analysis detects Covid-19 more reliably (11.12.2020)

A team of researchers at Inselspital, Bern University Hospital, and the ARTORG Center for Biomedical Research, University of Bern, has developed a new chest radiography image analysis for the detection of COVID-19. The algorithmic method was trained for various diagnoses on 8000 X-ray images. The researchers compared this artificial intelligence (AI) with standard, diagnostic annotation by radiologists. Especially for distinguishing Covid-19 from non-Covid-19 lung disease, AI provided significantly more reliable results.

Medienmitteilung, Insel Gruppe, 11 December 2020 (PDF, 213KB) Media release, Insel Gruppe, 11 December 2020 (PDF, 184KB)

AI to exploit the potential of the world’s fastest whole-body PET/CT scanner (02.12.2020)

Recently, the world’s fastest whole-body PET/CT scanner was installed at the Inselspital, Bern University Hospital. The scanner renders 4D images for a 106 cms field of view in just a few seconds, opening up new dimensions in research, diagnostics and therapy management. Kuangyu Shi, member of the Bern Center for Artificial Intelligence in Medicine (CAIM) and Head of Lab for AI and Translational Theranostics at the Department of Nuclear Medicine at Inselspital, explains how AI can help alleviate the challenges the new technology poses: "AI allows us to fully explore the machine’s potential. We can thus delve deeper into our research on AI and translational theranostics and customize clinical practice.”

AI for total body PET/CT (PDF, 92KB)

Bern founds Center for Artificial Intelligence in Medicine (17.11.2020)

The University of Bern and the Inselspital, Bern University Hospital, are founding a “Center for Artificial Intelligence in Medicine” (CAIM) that combines cutting-edge research, engineering and digitalization. Using artificial intelligence it will develop new medical technologies to enable tailor-made and efficient patient care. Partners of the new center are sitem-insel, the Swiss Institute for Translational and Entrepreneurial Medicine, and the University Psychiatry Services (UPD).

Medienmitteilung, Universität Bern, 17 November 2020 (PDF, 97KB) Media release, University of Bern, 17 November 2020 (PDF, 128KB) Communique de presse, Université de Berne, 17 novembre 2020 (PDF, 134KB)