Search

This contribution examines the concept of circular economy (CE) within the healthcare sector, introducing the 9R paradigm as a novel framework for the implementation of sustainable strategies. Given the significant CO2 emissions from the medical device sector and the heterogeneous digitalization across the industry, CE offers an avenue to minimize environmental impact while simultaneously enhancing patient care. Leveraging advanced data platforms and digital technology can help reduce resource waste and CO2 emissions, while also optimizing patient care by avoiding redundant diagnostics and treatments. The contribution provides a brief introduction for the application of the 9R paradigm in the healthcare sector and highlights the opportunities associated with the implementation of CE strategies. It outlines innovative data-driven business models based on CE and digital transformation principles. In conclusion, the strategic application of R-strategies can be described as a multi-layered approach requiring close collaboration among all stakeholders to shape a more sustainable and efficient future in healthcare and provide new digital business models.

Inpatient care facilities globally are facing a critical shortage of staff, posing significant challenges to resident well-being and care quality. This issue is further compounded by demographic shifts and increasing care demands. While technological advancements offer promise in alleviating nursing staff burdens, their effective integration remains complex, with nursing staff acceptance playing a pivotal role. This paper describes a systematic approach designed to streamline the process of identifying, categorizing, and prioritizing suitable technologies in inpatient care settings. By taking into account the specific needs and requirements of nursing staff, this approach, validated through a comprehensive case study, aims to facilitate targeted technology adoption, thereby contributing to the successful digitization of this occupational domain.

Background: Bladder cancer (BCa) is the second most common type of cancer in the genitourinary system and causes approximately 165,000 deaths each year. The diagnosis of BCa is primarily done through cystoscopy, which involves visually examining the bladder using an endoscope. Currently, white light cystoscopy is considered the most reliable method for diagnosis. However, it can be challenging to detect and diagnose flat, small, or poorly textured lesions. The study explores the performance of deep learning systems (YOLOv7- tiny, RT-DETR18), originally designed for detecting adenomas in colonoscopy images, when retrained and tested with cystoscopy images. The deep neural network used in the study was pre-trained on 35,699 colonoscopy images (some from Mannheim) and both architectures achieved a F1 score of 0.91 on publicly available colonoscopy datasets. Results: When the adenoma-detection network was tested with cystoscopy images from two sources (Ulm and Erlangen), F1 scores ranging from 0.58 to 0.81 were achieved. Subsequently, the networks were retrained using 12,066 cystoscopy images from Mannheim, resulting in improved F1 scores ranging from 0.77 to 0.85. Conclusion: It could be shown that transformer based networks perform slightly better than YOLOv7-tiny networks, but both network types are feasable for lesion detection in the human bladder. The retraining of the network with additional cystoscopy data led to an improvement in the performance of urinary lesion detection. This suggests that it is possible to achieve a domain-shift with the inclusion of appropriate additional data.

The integration of Artificial Intelligence (AI) into healthcare represents a transformative shift, offering opportunities for enhancing patient care, diagnostic accuracy, process optimization and treatment pathways. This research sets out to forge a strategic management decision support framework for leveraging AI within the healthcare sector, aimed at systematically exploring and integrating AI innovations to bolster the patient health outcomes. By creating a comprehensive categorization system, we attempt to navigate the complex array of possible AI applications within the field of healthcare, hence enabling the identification, selection, and advancement of AIdriven initiatives. Through a blend of systematic literature review and expert insights, this study maps possible AI applications across dimensions like ‘medical disciplines’, ‘healthcare processes’, ‘AI research areas’, and ‘user groups’. By reflecting the diverse perspectives, this system transcends mere classification and stands as a cornerstone for identifying, selecting, and developing AI-driven medical use cases to guide strategic implementations of AI within clinical settings. This multidimensional system offers a blueprint for healthcare entities to strategically navigate the AI landscape, enabling them to make informed decisions about technology adoption and change management processes, ultimately leading to improved patient care and operational efficiency.

The acquisition of two-dimensional X-ray images (of non-human objects) is quite expensive, primarily due to the costly flat-panel or line detectors. Reducing these costs could open up new possibilities in healthcare applications, such as inspecting limb prostheses for cracks and defects or safety-related material testing of helmets and child seats. In this contribution, two different setups of low-cost mono-pixel X-ray scanners are described: a small scanner with a working volume of 20 x 20 x 20 cm3 and a large system with a volume of 150 x 230 x 110 cm3. Using these novel mono-pixel X-ray scanning systems, various objects - ranging from an X-ray phantom and a human skull phantom to a high-end limb prosthesis with integrated electronics - were scanned and investigated. Compared to conventional X-ray systems, the mono-pixel systems can depict these objects without the distortion problems of cone beam detectors. Although the scanning procedure is time-consuming, the results are quite promising. Mono-pixel X-ray-scanning is a novel modality capable of scanning largescale objects with low-dose radiation. While it is currently not suitable for human applications due to the long scanning times, it can still be used for the non-destructive testing of large-scale medical devices, such as e.g. arm and leg prosthesis.

Background: As part of a open surgical intervention, it is desirable to identify, track and count the surgical instruments used, in order to document and potentially optimize the process. In the past, various technologies have been proposed to this end, as e.g. RFIDs attached to or included in the instruments, contact-less identification by image analysis or (bar-, QR-) codes, or counting by human resources, but all of them having their individual drawbacks. Objective: We present and evaluate a new method to identify metallic-conductive surgical instruments using Low Frequency (LF) magnetic fields. Method: The investigated LF magnetic field approach is based on the IndLoc system developed by the Fraunhofer IIS - originally for metallic asset identification within the field of supply chain management - and is applied on small collection of 13 typical surgical tools (four scissors, two needle holders, four tweezers, two retractors, a bone curette) for open surgery. Results: 9 out of the 13 instruments can be correctly identified, two pairs (2 short and 2 medium tweezers) are too self-similar and could not discriminated correctly from each other. Conclusion: These initial experiments show that an LF-based identification of metallic surgical tools is feasible, but the system needs to be modified (with more and smaller coils) to increase the geometrical resolution to better distinguish between self-similar and potentially more objects.

Background: The examination of lymph nodes (LNs) regarding metastases is vital for the staging of cancer patients, which is necessary for diagnosis and adequate treatment selection. Advancements in digital pathology, utilizing Whole-Slide Images (WSIs) and convolutional neural networks (CNNs), pose new opportunities to automate this procedure, thus reducing pathologists’ workload while simultaneously increasing the accuracy in metastases detection. Objective: To address the task of LN-metastases detection, the use of weakly supervised transformers are applied for the analysis of WSIs. Methods & Materials: As WSIs are too large to be processed as a whole, they are divided into non-overlapping patches, which are converted to feature vectors using a CNN network, pre-trained on HE-stained colon cancer resections. A subset of these patches serves as input for a transformer to predict if a LN contains a metastasis. Hence, selecting a representative subset is an important part of the pipeline. Hereby, a prototype based clustering is employed and different sampling strategies are tested. Finally, the chosen feature vectors are fed into a transformer-based multiple instance learning (MIL) architecture, classifying the LNs into healthy/negative (that is, containing no metastases), or metastatic/positive (that is, containing metastases). The proposed model is trained only on the Camelyon16 training data (LNs from breast cancer patients), and evaluated on the Camelyon16 test set. Results: The trained model achieves accuracies of up to 92.3% on the test data (from breast LNs). While the model struggles with smaller metastases, high specificities of up to 96.9% can be accomplished. Additionally, the model is evaluated on LNs from a different primary tumor (colon), where accuracies between 62.3% and 95.9% could be obtained. Conclusion: The investigated transformer-model performs very good on LN data from the public LN breast data, but the domain transfer to LNs from the colon needs more research.

Background: Due to an aging society and changing health behaviors, emergency room crowding has become a major problem in western health care systems. Empowering patients and health care workers to assess necessary and relevant information is critical to streamline clinical workflows. Health kiosks, designed for services like self-check-in or (ideally contactless) health self-assessment may be instrumental in solving this issue. Based on the collected data, automated workflows such as flagging critical patients, inducing specific diagnostics or early symptomatic treatment could be implemented. Objective: Using an AI-supported software, which visually analyzes and categorizes facial expressions, the emotional status of hemato-oncologic patients in a German oncology outpatient clinic was examined. Additionally a survey was conducted, evaluating the acceptance of such a self-assessment solution. Results: 98% of the participants were not stressed by the real-time emotion analysis. However, the current set of registered emotion categories was found to be only partially sufficient to adequately describe the emotional status of the patients. More importantly, 88% of the participants found such a system to be meaningful. Also, 84% of the participants agreed that such a self-analysis could be of potential assistance. No relevant generation- or gender-specific differences could be observed. Discussion: Automated analysis of patients’ emotional status can be a first step toward a more comprehensive assessment of the respective health status. Patients, in particular the elderly, approve to the vision and development of such a system. Next steps are a further improvement of the AI-based emotion recognition software with respect to more emotional states as well as the definition, inclusion and ideally contactless acquisition of physical biomarkers (as e.g. heart rate or respiratory rate) determining physical and mental well-being.

Background: Bladder cancer (BCa) is the second most common genitourinary malignancy and has a mortality of 165,000 deaths p.a. The diagnosis of BCa is mostly carried out using cystoscopy - the visual examination of the urinary bladder with an endoscope. White light cystoscopy is currently considered as gold standard for the diagnosis. Nevertheless, especially flat, small or weakly textured lesions, are very difficult to detect and diagnose. Objective: With the advent of deep learning and already commercially available systems for the detection of adenomas in colonoscopy, it is investigated how such a system - for colonoscopy - performs if retrained and tested with cystoscopy images. Methods: A deep neural network with a YOLOv7-tiny architecture was pre-trained on 35,699 colonoscopy images (partially from Mannheim), yielding a precision = 0.92, sensitivity = 0.90, F1 = 0.91 on public colonoscopy data collections. Results: Testing this adenomadetection network with cystoscopy images from three sources (Ulm, Erlangen, Pforzheim), F1 scores in the range of 0.67 to 0.74 could be achieved. The network was then retrained with 12,066 cystoscopy images (from Mannheim), yielding improved F1 scores in the range of 0.78 to 0.85. Conclusion: It could be shown that a deep learning network for adenoma detection in colonoscopy is ad-hoc able to detect approximately 75% of the lesions in the urinary bladder in cystoscopy images, suggesting that these lesions have a similar appearance. After retraining the network with additional cystoscopy data, the performance for urinary lesion detection could be improved, indicating that a domain-shift with adequate additional data is feasible.

Introduction: Application-related (hands-on) training and instruction is an essential component in the advanced education of physicians in residency. To teach and train required endoscopic examinations and interventions in a standardized manner, courses are offered by various societies. Within these courses the assistant endoscopists learn and train how to use the technical equipment and how to perform the examinations and related interventions. Similarly, for graduate biomedical engineering students with a job- or research-related interest in the field of endoand laparoscopy, an early involvement and practical introduction to the instrumentation and techniques - if possible already during their studies - can generate a fundamental understanding of the technical requirements as well needs of the physicians. Objective: As various such (hands-on) training-courses for flexible and rigid endoscopy specifically addressing young physicians, surgeons as well as engineers are offered by numerous institutes, this contribution tries do provide a structured overview on/over these possibilities. Material: Known cross-institutional courses in the field of endo- and laparoscopy currently offered in Germany in cooperation of physicians and engineers are listed and briefly described. Results: A total of n = 4 crossinstitutional courses for the introduction and training in flexible and rigid endoscopy have been identified. Discussion: The cross-institutional courses for physicians and technicians in the field of endo- and laparoscopy presented serve to provide a better understanding of the subject matter and the field of either discipline. Questions that arise in the field of endoscopy from physicians can be addressed at an early stage and, if necessary, worked on with little effort. Ideas for new instruments and even new interventions can arise and be pursued through an intensive exchange between technicians and physicians. Collaborations are established on a local basis which will foster national activities in this field and which allow to develop competencies relevant for the industrial sector of medical engineering.

With a growing number of people in need of care and the shortage of skilled workers, caregivers express the desire for improved support to reduce physical strain and psychological stress. One task for burden relieve is documentation of nursing actions and treatment processes, as this can be done in parallel with the action. Hence tools capturing information in real time could improve information sharing. Also, contactless assessment of patients’ vital signs and emotions is desirable. Thus, the idea is to enhance an existing mobile platform, which can follow the nurse, includes a voice-controlled assistant, and be able for wireless (e.g., via Bluetooth) assessment of vital data from patients in its direct vicinity. Using the Temi telepresence robot as an open technical platform, we provide and discus requirements for this platform to potentially aid and relief to medical staff within caring and elderly homes, with respect to voice-control, Bluetooth-based vital data collection, interfaces to a digital patient manager and automated emotion recognition. The presented concept and the rating of the associated technologies leads to the conclusion that - if adequately implemented and installed - the caregivers could potentially receive noticeable relief from them. Nevertheless, as next steps, the examined technologies and ideas must be completely implemented, thoroughly fused with each other, and evaluated to provide a real support system for clinical care and assisted documentation.

We explore the task of tissue classification for colon cancer histology in a low label regime comparing a semi-supervised and a supervised learning strategy in a series of experiments. Further, we investigate the model robustness w.r.t. distribution shifts in the unlabeled data and domain shifts across different scanners to prove their practicality in a histology context. By utilizing unlabeled data in addition to nl = 1000 labeled tiles per class, we yield a substantial increase in accuracy from 89.9% to 91.4%.

Introduction: In the field of colonoscopy results of automated or supporting procedures using medical image processing are often compromised by contaminants such as blood or stool, but also due to medical instruments. Such objects or contaminations often obscure relevant areas or cause disturbances in various algorithms and also prevent physicians from recognizing relevant information in the video-stream. Hence, automatic detection and segmentation of such contaminations is desirable, by obtaining proper segmentation masks to an algorithm yields the possibility to exclude such areas from analysis. Method: We created a training and validation dataset by collecting (and annotating, where necessary) endoscopic video sequences from various public available datasets and a privat collection. An encoder-decoder deep convolutional neural network was trained with this data to predict segmentation masks for blood, stool, medical instruments and other objects. Results: Depending on the class (blood, stool, ... ), we were able to obtain a mean Dice score in the range between 0.88 to 0.93 on the evaluation dataset. Conclusion: The proposed approach hence allows us providing segmentation masks that are able to reduce problems in the subsequent image analysis and also find regions that were not properly observed.

Motivation: Within the field of bio-implants one topic deals with the in-vitro production of 3D tissues from cell cultures. A major challenge in the cultivation of biopreparations is the imitation of the vascular system with its surrounding tissue. Additiv manufacturing of such vascularisation could be supported with an adequate template from high-resolution 3D radiographs. Goals: Evaluation of highresolution 3D-imaging methods and 3D-analysis approaches to extract the vascularisation from tissue samples. Methods: To provide such template, vascular structure of small tissues samples (partially treated with contrast medium) were recorded and compared using high-resolution CT and MRI imaging modalities. Optimal measurement parameters were selected for the acquisition of very small vessels. Interactive and semi-automated segmentation of the vessel system were investigated and compared. Results: MRI scans yield a higher contrast than CT scans, but are much slower. Duration of interactive segmentation ranges between 1 to 12 hours. Runtime of the (semi) automatic method was between 5 and 20 minutes, not counting manual adjustment of the parameters. Correlation between manual and automatic segmentation yield Hausdorff distances of 0.024 (CT) and 0.74 (MRI) and Dice coefficients of 0.7 (CT) and 0.39 (MRI). Conclusion: Both imaging methods are appropriate for high-resolution vessel detection and segmentation, nevertheless, MRI with no contrast agent seems preferable if the imaging time can be reduced.

Deep Learning-based tissue classification may support pathologists in analyzing digitized whole slide images. However, in such critical tasks, only approaches that can be validated by medical experts in advance to deployment, are suitable. We present an approach that contributes to making automated tissue classification more transparent. We step beyond broadly used visualizations for last layers of a convolutional neural network by identifying most relevant intermediate layers applying Grad-CAM. A visual evaluation by a pathologist shows that these layers assign relevance, where important morphological structures are present in case of correct class decisions. We introduce a tool that can be easily used by medical experts for such validation purposes for any convolutional neural network and any layer. Visual explanations for intermediate layers provide insights into a neural network’s decision for histopathological tissue classification. In future research also the context of the input data must be considered.

Introduction: Early detection of adenomas and polyps is one central goal of colonoscopic screening programs. As the adenoma detection rate (ADR) depends on the experience of the endoscopist, AI-based polyp detection systems can be used for real-time assistance. Hence, to support the physicians such AI-based systems using deep-convolutional neural networks (DCNNS) have been introduced in the past years. One disadvantage of these techniques is the need of a huge amount of labeled training data. Method: We investigate a "human-in-the-loop approach" to minimize the required time to generate labeled training data. The approach is evaluated within the training a YOLOv4 neural network to detect polyps in colonoscopic image data. The performance metrics of the neural network are evaluated on three public datasets. Results: The performance of the YOLO network increased from a precision of 0.88, recall of 0.83, F1 score of 0.86, and a F2 score of 0.86 to a precision of 0.91, recall of 0.87, F1 = of 0.89, and F2 = of 0.88. The interactive labelling of 1,000 images only takes one hours. Conclusion: The proposed "human-in-theloop approach" is capable of generating labelled image data in a minimum of time while increasing the performance metrics as well. For higher performance increase more data can now be labeled within this new approach.

When documenting colonoscopic examinations, the challenge arises, that the exact location of the detected lesions (such as polyps) can usually only be determined with low precision. In order to improve the image-based documentation of such lesions with respect to their location and context, a system is currently investigated and developed, allowing the creation of a panoramic map of the interior of the colon’s lumen during an endoscopic examination. To evaluate the proposed algorithms, a 3D-printed colon phantom with known geometry has been developed. Furthermore, to obtain image sequences with precise and known motion during a simulated examination for test purposes, a robot was used to perform the insertion and withdrawal of the endoscope with constant speed and a steady viewing angle. The thus obtained image data is used to generate an ideal panoramic map using our the proposed stitching algorithm. Hence, it is possible to obtain reference maps of the phantom, which can be used to evaluate the quality of the algorithm.

For the development, training, and validation of AIbased procedures, such as the analysis of clinical data, prediction of critical events, or planning of healthcare procedures, a lot of data is needed. In addition to this data of any origin (image data, bio-signals, health records, machine states, …) adequate supplementary information about the meaning encoded in the data is required. With this additional information - the semantic or knowledge - a tight relation between the raw data and the human-understandable concepts from the real world can be established. Nevertheless, as the amount of data needed to develop robust AI-based methods is strongly increasing, the assessment and acquisition of the related knowledge becomes more and more challenging. Within this work, an overview of currently available concepts of knowledge acquisition are described and evaluated. Four main groups of knowledge acquisition related to AI-based technologies have been identified. For image data mainly iconic annotation methods are used, where experienced users mark or draw depicted entities in the images and label them using predefined sets of classifications. Similarly, bio-signals are manually labelled, whereby important events along the timeline are marked. If no sufficient data is available, augmentation and simulation techniques are applied yielding data and semantics at the same time. In applications, where expensive sensors are replaced by low-cost devices, the high-grade data can be used as semantics. Finally, classic rule-based approaches are used, where human factual and procedural knowledge about the data and its context is translated into machine-understandable procedures. All these methods are depending on the involvement of human experts. To reduce this, more intelligent and hybrid approaches are needed, shifting the focus from the-human-in-the-loop to the-machine- in-the-loop.

Multispectral imaging devices incorporating up to 256 different spectral channels have recently become available for various healthcare applications, as e.g. laparoscopy, gastroscopy, dermatology or perfusion imaging for wound analysis. Currently, the use of such devices is limited due to very high investment costs and slow capture times. To compensate these shortcomings, single sensors with spectral masking on the pixel level have been proposed. Hence, adequate spectral reconstruction methods are needed. Within this work, two deep convolutional neural networks (DCNN) architectures for spectral image reconstruction from single sensors are compared with each other. Training of the networks is based on a huge collection of different MSI imagestacks, which have been subsampled, simulating 16-channel single sensors with spectral masking. We define a training, validation and test set (‘HITgoC’) resulting in 351 training (631.128 sub-images), 99 validation (163.272 sub-images) and 51 test images. For the application in the field of neurosurgery an additional testing set of 36 image stacks from the Nimbus data collection is used, depicting MSI brain data during open surgery. Two DCNN architectures were compared to bilinear interpolation (BI) and an intensity difference (ID) algorithm. The DCNNs (ResNet-Shinoda) were trained on HITgoC and consist of a preprocessing step using BI or ID and a refinement part using a ResNet structure. Similarity measures used were PSNR, SSIM and MSE between predicted and reference images. We calculated the similarity measures for HitgoC and Nimbus data and determined differences of the mean similarity measure values achieved with the ResNet-ID and baseline algorithms such as BI algorithm and ResNet-Shinoda. The proposed method achieved better results against BI in SSIM (.0644 vs. .0252), PSNR (15.3 dB vs. 9.1 dB) and 1-MSE*100 (.0855 vs. .0273) and compared to ResNet-Shinoda in SSIM (.0103 vs. .0074), PSNR (3.8 dB vs. 3.6 dB) and 1-MSE*100 (.0075 vs. .0047) for HITgoC/Nimbus. In this study, significantly better results for spectral reconstruction in MSI images of open neurosurgery was achieved using a combination of ID-interpolation and ResNet structure compared to standard methods.

For the image-based documentation of a colonoscopy procedure, a 3D-reconstuction of the hollow colon structure from endoscopic video streams is desirable. To obtain this reconstruction, 3D information about the colon has to be extracted from monocular colonoscopy image sequences. This information can be provided by estimating depth through shape-from-motion approaches, using the image information from two successive image frames and the exact knowledge of their disparity. Nevertheless, during a standard colonoscopy the spatial offset between successive frames is continuously changing. Thus, in this work deep convolutional neural networks (DCNNs) are applied in order to obtain piecewise depth maps and point clouds of the colon. These pieces can then be fused for a partial 3D reconstruction.

Robot-based service platforms are currently establishing themselves as new and affordable variants for supporting care in elderly, retirement and nursing homes. Many are open multifunctional platforms, which can potentially be integrated into such environments, if the necessary infrastructure is available. Furthermore, many services can be realized on these platforms, which can be used to foster distant interactions between inhabitants and care-providers, while simultaneously keeping up the quality of life of the inhabitants. Open mobile robotic platforms allow the extension with adequate new sensors. To detect infectious diseases of residents and healthcare-professionals, optical sensors can be used for the assessment of vital data such as heartrate and heartrate variability, respiratory rate, SpO2 or temperature. Additionally, you can consider demographic data (age, gender, constitution) of the observed person for the optical assessment, i.e. obtained by facial image analysis. As these mobile platforms are also equipped for telepresence, in case of detecting an infected person, these systems support video conferencing with their built-in cameras and microphones. Finally, the interaction with the electronic care record is necessary to upload all acquired vital data and further relevant information. All the named technologies have been under investigation in the past years and are currently moving from laboratory settings to real-world scenarios. Nevertheless, the smooth integration of all components into one system architecture in combination with (AI-based) data analysis are still open issues.

Infective endocarditis (IE) is an infection of the endocardium, and the heart valves associated with high morbidity and mortality. Fluorescence in situ Hybridization (FISH) is a molecular imaging technique used for diagnosis of IE based on histological heart valve tissue sections. FISH allows detection and identification of microorganisms and gives information about their quantity and spatial distribution. This information is important to guide appropriate antibiotic treatment. However, as manual FISH image analysis is time- and costexpensive, an automated image analysis pipeline (consisting of tissue segmentation, bacteria detection, and spot detection modules) is proposed to assist locating potential regions with microorganisms. The proposed approach was evaluated in a study, where five observers manually assessed a set of 171 fields-of-view (FoVs) captured in 400-fold magnification from 10 randomly chosen WSI for the presence of microorganisms, morphologically detected by the nucleic acid stain DAPI. The task of the observers was to mark the presented image using a 2-class score (‘positive/questionable’ or ‘negative’). The human assessment was compared to the results suggested by the algorithm. The proposed algorithm locates and ranks potential regions with microorganisms in heart valve sections so that experts can validate them in higher power FoVs for the presence of bacteria and identify their species. The automated system for preselecting and recommending adequate FoVs is thus a starting point to support experts and save human resources. It is now ready to be further developed for the detection of bacteria by FISH.

Currently, various AI-based systems for computerassisted adenoma- and polyp-detection during colonoscopy have been brought to the market and are under clinical investigation. With these systems available to be used during routine screening colonoscopy and first results published about experiments and findings, it has become of interest how and to which extend such systems are used during the examination. Specifically, similarly to automotive navigation, it is of interest of how much visual focus is put onto the augmented image of the above-mentioned devices, signalling possible hypothesis of adenomas or polyps, and how much time-of-attention remains on the original colonoscopic video data. Thus, within a study, N = 36 participants using a prototype of a polypdetection system have been observed with an eye-tracker-system, to capture and evaluate the relative time of attention with respect to the original and augmented video data and differentiate these values between various sub-groups based on experience, education and gender. T-tests were conducted to identify potential significant differences. Based on the obtained data, the augmented video data is used with a very high attention (up to 75%) depending on the regarded sub-group. Experienced as well as less-experienced users (with > 500 colonoscopies) both preferred looking at the original data. In contrast, gastroenterologists (in contrast to nurses, students, engineers) were more interested in the outcome of the novel AIsystem. The female group preferred looking at the unobstructed data, while the male group was highly interested in the AI-based data.

The goal of our research work is the development of a novel endoscopic anastomosis device for the colon. One of the main challenges in this context is the application of forces at the endoscope tip to rejoin the two bowel endings. Thus, we focus on a magnetic two-part compression implant approach. The implant halves are detached from the applicator units by means of electromagnets. In this contribution we present the results of our experiments to determine the implant design with special focus on tissue compression forces and the resultant electromagnet dimensioning to estimate size requirements of the application/detachment system. To achieve the targeted compression forces derived from literature, we used cubic N52 magnetized neodymium magnets1 with a side length of 5 mm and mild steel screws. For these magnets, we evaluated a required electromagnetic repulsion force of 4.1 N. For the electromagnetic detachment system this led to the need for 166 windings for the coils on oral side, and 146 windings for the coils at the aboral side. Based on these requirements, a colonoscope diameter (~14 mm) increase of 10.6 mm on the oral side and of 12 mm on the aboral side due to the application device must be assumed. Nevertheless, this diameter still remains within the size range of other colonoscopic tools, such as e.g., circular staplers.

Since 2021, the Fraunhofer Institutes ITEM and IIS and the German Society for Biomedical Engineering (VDE DGBMT) have launched a new online seminary. This sustainable format, referred to as "Expert Panel", stimulates scientific interaction by introducing research topics and innovations within sixty minutes. The panel with three experts from different fields presents and discusses perspectives from science, clinical practice, and the industry. The first season of the "Expert Panel" offered comprehensive insights on the latest developments and innovations within "New Interventional Methods for Tomorrow's Hearing". Each topic was presented by a triad of experts from science, clinical practice, as well as the industry. The sessions comprised life-demonstration of planning tools for surgical interventions, the presentation of patient-individual intervention techniques and novel audiological systems for Cochlea Implant (CI) fitting and many more inspiring technologies. Each “Expert Panel” concluded with a moderated Q&A final. The sessions provided condensed updates and unique insights from applied research and had a lively attendance of up to more than 120 participants from the fields of research, clinic, and industry. This kind of format does not only drive interpersonal and interdisciplinary knowledge transfer - especially in times of home office - it also enables establishing a community of network partners, who meet at the “Expert Panel” on a regular basis. In addition, education credits have been offered to the medical community, thus creating additional benefit for the participants.

For the gastroscopic examination of the stomach, the restricted field of view related to the „keyhole“-perspective of the endoscope is known to be a visual limitation. Thus, a panoramic extension can enlarge the field of vision, supports the endoscopist during the examination, and ensures that all of the inner stomach walls are visually inspected. To compute such a panorama of the stomach, knowledge about the geometry of the underlying structure is required. Structure from motion an approach to reconstruct the necessary information about the 3D-structure from monocular image sequences as provided by a gastroscope. We examine and evaluate an existing deep neuronal network for stereo reconstruction, in order to approximate the geometry of stomach parts from a set of consecutive acquired image pairs from gastroscopic videos.

All aspects of daily life increasingly include digitization. So-called „smart home“ technologies, as well as „wearables“, are gaining attention from more and more dwellers. Therefore, sensor-based, individualized, AI-based services for improved post-intervention monitoring and therapy accompaniment will become feasible and possible if these systems offer a related context-awareness. This paper provides an approach on how to sense and interpret specific contexts with the help of wearables, smartwatches, smart home sensors, and emotion detection software.

Early detection of polyps is one central goal of colonoscopic screening programs. To support gastroenterologists during this examination process, deep convolutional neural network can be applied for computer-assisted detection of neoplastic lesions. In this work, a Mask R-CNN architecture was applied. For training and testing, three independent colonoscopy data sets were used, including 2484 HD labelled images with polyps from our clinic, as well as two public image data sets from the MICCAI 2015 polyp detection challenge, consisting of 612 SD and 194 HD labelled images with polyps. After training the deep neural network, best results for the three test data sets were achieved in the range of recall = 0.92, precision = 0.86, F1 = 0.89 (data set A), rec = 0.86, prec = 0.80, F1 = 0.82 (data set B) and rec = 0.83, prec = 0.74, F1 = 0.79 (data set C).

Cochlea Implant (CI) planning is usually based on preoperative obtained CT or MRI data, visualising risk structures in the petrosal bone. In the past years, Digital Volume Tomography (DVT) with an increased spatial resolution and reduced radiation has become more important in the clinical routine for otology. In this work we propose an extension of our interactive “wizard”-guided approach for the interactive segmentation of the middle and inner ear structures for the use with DVT data. Different filter pipelines enable the user to interactive segment the acoustic canal, ossicles, tympanic cavity, facial nerve, chorda tympani, round window, cochlea and semicircular canals. The approach has been evaluated on six pre-operative acquired DVT datasets by an ENT expert. Results imply that the proposed method can handle DVT well and can potentially be used for interactive OR planning.

Dual energy radiographic imaging is a method to provide material information and can be used to differentiate between various tissue types. Dual energy X-ray absorption (DEXA) can be applied for breast density, osteoporosis or bone fracture analysis. To support radiologists with the assessment of DEXA images, machine learning can be applied. Specifically, deep convolutional neural networks (DCNNs) can be used for medical image analysis. In this work a DCNN is proposed and evaluated for automated detection of bone splinters in DEXA phantom images. The image data consists of 47 phantoms with (35) and without (12) bone splinters. Material decomposition and energy weighting results in additional image channels. Various DCNN architectures and parameters were explored. A classification rate in regions with 90 % and without 99 % bone splinters was achieved.

Automated image processing methods enable objective, reproducible and high quality analysis of fluorescent cell images in a reasonable amount of time. Therefore, we propose the application of image processing pipelines based on established segmentation algorithms which can handle massive amounts of whole slide imaging data of multiple fluorescent labeled cells. After automated parameter adaption the segmentation pipelines provide high quality cell delineations revealing significant differences in the spreading of B cells: LPS-activated B cells spread significantly less on anti CD19 mAb than on anti BCR mAb and both processes could be inhibited by the F-actin destabilizing drug Cytochalasin D. Moreover, anti CD19 mAb induce a more symmetrical spreading than anti BCR mAb as reflected by the higher cell circularity.

This paper summarizes our initial efforts to reconstruct the urinary bladder from endoscopic images acquired in the clinical routine. We found that up to now, only very few attempts have been reported which achieve a true 3D reconstruction of the human bladder. One promising approach which yields a geometric reconstruction up to scale from a monocular stream of images is highlighted and our initial results obtained from adapting the method for its use in clinical cystoscopy are presented.

Abstract We present a novel solution for automatic surface inspection of metallic tubes by applying a structured illumination. The strength of the proposed approach is that both structural and textural surface defects can be visually enhanced, detected, and well separated from acceptable surfaces. We propose a machine vision approach and we demonstrate that this technique is applicable in an industrial setting. We show that recording artefacts drastically increases the complexity of the inspection task. The algorithm implemented in the industrial application and which permits the segmentation and classification of surface defects is briefly described. The suggested method uses "perturbations from the stripe illumination" to detect, segment, and classify any defects. We emphasize the robustness of the algorithm against recording artefacts. Furthermore, this method is applied in 24 h/7 day real-time industrial surface inspection system.