Lebenswissenschaften und Ethik
Refine
Year of publication
Document Type
Language
- English (285) (remove)
Is part of the Bibliography
- no (285)
Keywords
- Deep Learning (13)
- Artificial Intelligence (12)
- Diagnose (11)
- Maschinelles Lernen (9)
- Bildgebendes Verfahren (8)
- Humans (8)
- Künstliche Intelligenz (7)
- Simulation (7)
- Speiseröhrenkrankheit (7)
- Biomechanics (6)
Institute
- Fakultät Informatik und Mathematik (105)
- Fakultät Maschinenbau (93)
- Regensburg Center of Biomedical Engineering - RCBE (88)
- Regensburg Medical Image Computing (ReMIC) (70)
- Regensburg Center of Health Sciences and Technology - RCHST (43)
- Fakultät Angewandte Sozial- und Gesundheitswissenschaften (42)
- Labor Biomechanik (LBM) (42)
- Fakultät Elektro- und Informationstechnik (19)
- Fakultät Betriebswirtschaft (15)
- Institut für Sozialforschung und Technikfolgenabschätzung (IST) (15)
Begutachtungsstatus
- peer-reviewed (285) (remove)
Background: For surgical fixation of bone fractures of the human hand, so-called Kirschner-wires (K-wires) are drilled through bone fragments. Due to the minimally invasive drilling procedures without a view of risk structures like vessels and nerves, a thorough training of young surgeons is necessary. For the development of a virtual reality (VR) based training system, a three-dimensional (3D) printed phantom hand is required. To ensure an intuitive operation, this phantom hand has to be realistic in both, its position relative to the driller as well as in its haptic features. The softest 3D printing material available on the market, however, is too hard to imitate human soft tissue. Therefore, a support-material (SUP) filled metamaterial is used to soften the raw material. Realistic haptic features are important to palpate protrusions of the bone to determine the drilling starting point and angle. An optical real-time tracking is used to transfer position and rotation to the training system.
Methods: A metamaterial already developed in previous work is further improved by use of a new unit cell. Thus, the amount of SUP within the volume can be increased and the tissue is softened further. In addition, the human anatomy is transferred to the entire hand model. A subcutaneous fat layer and penetration of air through pores into the volume simulate shiftability of skin layers. For optical tracking, a rotationally symmetrical marker attached to the phantom hand with corresponding reference marker is developed. In order to ensure trouble-free position transmission, various types of marker point applications are tested.
Results: Several cuboid and forearm sample prints lead to a final 30 centimeter long hand model. The whole haptic phantom could be printed faultless within about 17 hours. The metamaterial consisting of the new unit cell results in an increased SUP share of 4.32%. Validated by an expert surgeon study, this allows in combination with a displacement of the uppermost skin layer a good palpability of the bones. Tracking of the hand marker in dodecahedron design works trouble-free in conjunction with a reference marker attached to the worktop of the training system.
Conclusions: In this work, an optically tracked and haptically correct phantom hand was developed using dual-material 3D printing, which can be easily integrated into a surgical training system.
The elite sport movement for athletes with hearing impairments, namely Deaflympics, differs from the Paralympic and Olympic sport movements because it exhibits a variety of distinct sociocultural and organisational characteristics. Yet, mental training with Deaflympic athletes receives little to no attention from the scientific community. Little is known about sport psychology consultants’ (SPCs) work with Deaflympic athletes. In this study, we explored SPCs’ exposure to so called Deaf sport as well as their experiences, attitudes, and assumptions regarding the utility of psychological skills training (PST) with Deaflympic athletes. A self-constructed questionnaire with closed and semi-open questions was completed by 93 (58.8% female) SPCs in European German-speaking countries. Analyses revealed SPCs had limited exposure to Deaflympic sport but indicated readiness to work with Deaflympic athletes. SPCs shared no reasons as to why PST skills and techniques would not be effective with Deaflympic athletes. However, SPCs regarded communication challenges as a major obstacle. We conclude that the integration of elite Deaf sport in SPCs’ training programmes is vital, considering SPCs’ lack of exposure and experience with Deaflympic athletes as well as their communication insecurities. In addition, further empirical research on PST effectiveness in Deaflympics athletes should provide the foundation for evidence-based utility of applied sport psychology in Deaflympic sport.
Background
This study evaluated the effect of an artificial intelligence (AI)-based clinical decision support system on the performance and diagnostic confidence of endoscopists in their assessment of Barrett’s esophagus (BE).
Methods
96 standardized endoscopy videos were assessed by 22 endoscopists with varying degrees of BE experience from 12 centers. Assessment was randomized into two video sets: group A (review first without AI and second with AI) and group B (review first with AI and second without AI). Endoscopists were required to evaluate each video for the presence of Barrett’s esophagus-related neoplasia (BERN) and then decide on a spot for a targeted biopsy. After the second assessment, they were allowed to change their clinical decision and confidence level.
Results
AI had a stand-alone sensitivity, specificity, and accuracy of 92.2%, 68.9%, and 81.3%, respectively. Without AI, BE experts had an overall sensitivity, specificity, and accuracy of 83.3%, 58.1%, and 71.5%, respectively. With AI, BE nonexperts showed a significant improvement in sensitivity and specificity when videos were assessed a second time with AI (sensitivity 69.8% [95%CI 65.2%–74.2%] to 78.0% [95%CI 74.0%–82.0%]; specificity 67.3% [95%CI 62.5%–72.2%] to 72.7% [95%CI 68.2%–77.3%]). In addition, the diagnostic confidence of BE nonexperts improved significantly with AI.
Conclusion
BE nonexperts benefitted significantly from additional AI. BE experts and nonexperts remained significantly below the stand-alone performance of AI, suggesting that there may be other factors influencing endoscopists’ decisions to follow or discard AI advice.
Aims
Evaluation of the add-on effect an artificial intelligence (AI) based clinical decision support system has on the performance of endoscopists with different degrees of expertise in the field of Barrett's esophagus (BE) and Barrett's esophagus-related neoplasia (BERN).
Methods
The support system is based on a multi-task deep learning model trained to solve a segmentation and several classification tasks. The training approach represents an extension of the ECMT semi-supervised learning algorithm. The complete system evaluates a decision tree between estimated motion, classification, segmentation, and temporal constraints, to decide when and how the prediction is highlighted to the observer. In our current study, ninety-six video cases of patients with BE and BERN were prospectively collected and assessed by Barrett's specialists and non-specialists. All video cases were evaluated twice – with and without AI assistance. The order of appearance, either with or without AI support, was assigned randomly. Participants were asked to detect and characterize regions of dysplasia or early neoplasia within the video sequences.
Results
Standalone sensitivity, specificity, and accuracy of the AI system were 92.16%, 68.89%, and 81.25%, respectively. Mean sensitivity, specificity, and accuracy of expert endoscopists without AI support were 83,33%, 58,20%, and 71,48 %, respectively. Gastroenterologists without Barrett's expertise but with AI support had a comparable performance with a mean sensitivity, specificity, and accuracy of 76,63%, 65,35%, and 71,36%, respectively.
Conclusions
Non-Barrett's experts with AI support had a similar performance as experts in a video-based study.
Aims
Barrett´s esophagus related neoplasia (BERN) is difficult to detect and characterize during endoscopy, even for expert endoscopists. We aimed to assess the add-on effect of an Artificial Intelligence (AI) algorithm (Barrett-Ampel) as a decision support system (DSS) for non-expert endoscopists in the evaluation of Barrett’s esophagus (BE) and BERN.
Methods
Twelve videos with multimodal imaging white light (WL), narrow-band imaging (NBI), texture and color enhanced imaging (TXI) of histologically confirmed BE and BERN were assessed by expert and non-expert endoscopists. For each video, endoscopists were asked to identify the area of BERN and decide on the biopsy spot. Videos were assessed by the AI algorithm and regions of BERN were highlighted in real-time by a transparent overlay. Finally, endoscopists were shown the AI videos and asked to either confirm or change their initial decision based on the AI support.
Results
Barrett-Ampel correctly identified all areas of BERN, irrespective of the imaging modality (WL, NBI, TXI), but misinterpreted two inflammatory lesions (Accuracy=75%). Expert endoscopists had a similar performance (Accuracy=70,8%), while non-experts had an accuracy of 58.3%. When AI was implemented as a DSS, non-expert endoscopists improved their diagnostic accuracy to 75%.
Conclusions
AI may have the potential to support non-expert endoscopists in the assessment of videos of BE and BERN. Limitations of this study include the low number of videos used. Randomized clinical trials in a real-life setting should be performed to confirm these results.
Regarding the prevention of injuries and rehabilitation of the human hand, musculoskeletal simulations using an inverse dynamics approach allow for insights of the muscle recruitment and thus acting forces on the hand. Currently, several hand models from various research groups are in use, which are mainly validated by the comparison of numerical and anatomical moment arms. In contrast to this validation and model-building technique by cadaver studies, the aim of this study is to further validate a recently published hand model [1] by analyzing numerically calculated muscle activities in comparison to experimentally measured electromyographical signals of the muscles. Therefore, the electromyographical signals of 10 hand muscles of five test subjects performing seven different hand movements were measured. The kinematics of these tasks were used as input for the hand model, and the numerical muscle activities were computed. To analyze the relationship between simulated and measured activities, the time difference of the muscle on- and off-set points was calculated, which resulted in a mean on- and off-set time difference of 0.58 s between the experimental data and the model. The largest differences were detected for movements that mainly addressed the wrist. One major issue comparing simulated and measured muscle activities of the hand is cross-talk. Nevertheless, the results show that the hand model fits the experiment quite accurately despite some limitations and is a further step toward patient-specific modeling of the upper extremity.
Musculoskeletal lower back load of accoucheurs during childbirth – A pilot and feasibility study
(2021)
Introduction: Back problems represent one of the leading causes of accouchers' work-related musculoskeletal morbidities. The correct execution of birth-related maneuvers including manual perineal protection is crucial not only for the mother and child but also for obstetricians and midwives to reduce any strain on their musculoskeletal system. Therefore, the overall aim of this study was to test the feasibility of determining the effect of different accouchers' postures (standing and kneeling) on their musculoskeletal system.
Methods: The biomechanical analysis is based on musculoskeletal simulations that included motion recordings of real deliveries as well as deliveries conducted on a birthing simulator. These simulations were then used to determine individual joints' loads.
Results: In the kneeling posture, both a low intra-operator variability and a lower average maximum load of the lower back was observed. For the standing position the spine load was reduced by pivoting the elbow on the accouchers' thigh, which in turn was associated with a significantly greater load on the shoulder joint.
Conclusion: The study demonstrated the feasibility of our technique to assess joints loads. It also provided initial data indicating that a posture that reduces spinal flexion and tilt, achieved in this study by the kneeling, can significantly reduce the strain on the practitioner's musculoskeletal system.
Elbow stability is derived from a combination of muscular, ligamentous, and bony structures. After an elbow trauma the stability of the joint is an important decision criterion for the subsequent treatment. The decision regarding non-operative/operative care depends mostly on subjective assessments of medical experts. Therefore, the aim of this study is to use musculoskeletal simulations as an objective assessment tool to investigate the extent to which failure of different stabilizers affects the elbow stability and how these observations correspond to the assessment from clinical practice. A musculoskeletal elbow simulation model was developed for this aim. To investigate the stability of the elbow, varus/valgus moments were applied under 0°, 45°and 90° flexion while the respective cubital angle was analyzed. This was performed for nine different injury scenarios, which were also evaluated for stability by clinical experts. With the results, it can be determined by which injury pattern and under which flexion angle the elbow stability is impaired regarding varus/valgus moments. The scenario with a complete failure of the medial and lateral ligaments and a fracture of the radial head was identified as having the greatest instability. The study presented a numerical determination of elbow stability against varus/valgus moments regarding clinical injury patterns, as well as a comparison of the numerical outcome with experience gained in clinical practice. The numerical predictions agree well with the assessments of the clinical specialists. Thus, the results from musculoskeletal simulation can make an important contribution to a more objective assessment of the elbow stability.
Outputs of musculoskeletal models should be considered probabilistic rather than deterministic as they are affected by inaccuracies and estimations associated with the development of the model. One of these uncertainties being critical for modeling arises from the determination of the muscles' line of action and the physiological cross-sectional area. Therefore, the aim of this study was to evaluate the outcome sensitivity of model predictions from a musculoskeletal hand model in comparison to the uncertainty of these input parameters. For this purpose, the kinematics and muscle activities of different hand movements (abduction of the fingers, abduction of the thumb, and flexion of the thumb) were recorded. One thousand simulations were calculated for each movement using the Latin hypercube sampling method with a corresponding variation of the muscle origin/insertion points and the cross-sectional area. Comparing the standard hand to simulations incorporating uncertainties of input parameters shows no major deviations in on- and off-set time point of muscle activities. About 60% of simulations are located within a ± 30% interval around the standard model concerning joint reaction forces. The comparison with the variation of the input data leads to the conclusion that the standard hand model is able to provide not over-scattered outcomes and, therefore, can be considered relatively stable. These results are of practical importance to the personalization of a musculoskeletal model with subject-specific bone geometries and hence changed muscle line of action.
We propose an automatic approach for early detection of adenocarcinoma in the esophagus. High-definition endoscopic images (50 cancer, 50 Barrett) are partitioned into a dataset containing approximately equal amounts of patches showing cancerous and non-cancerous regions. A deep convolutional neural network is adapted to the data using a transfer learning approach. The final classification of an image is determined by at least one patch, for which the probability being a cancer patch exceeds a given threshold. The model was evaluated with leave one patient out cross-validation. With sensitivity and specificity of 0.94 and 0.88, respectively, our findings improve recently published results on the same image data base considerably. Furthermore, the visualization of the class probabilities of each individual patch indicates, that our approach might be extensible to the segmentation domain.
We investigate contrastive learning in a multi-task learning setting classifying and segmenting early Barrett’s cancer. How can contrastive learning be applied in a domain with few classes and low inter-class and inter-sample variance, potentially enabling image retrieval or image attribution? We introduce a data sampling strategy that mines per-lesion data for positive samples and keeps a queue of the recent projections as negative samples. We propose a masking strategy for the NT-Xent loss that keeps the negative set pure and removes samples from the same lesion. We show cohesion and uniqueness improvements of the proposed method in feature space. The introduction of the auxiliary objective does not affect the performance but adds the ability to indicate similarity between lesions. Therefore, the approach could enable downstream auto-documentation tasks on homogeneous medical image data.
Semantic segmentation is an essential task in medical imaging research. Many powerful deep-learning-based approaches can be employed for this problem, but they are dependent on the availability of an expansive labeled dataset. In this work, we augment such supervised segmentation models to be suitable for learning from unlabeled data. Our semi-supervised approach, termed Error-Correcting Mean-Teacher, uses an exponential moving average model like the original Mean Teacher but introduces our new paradigm of error correction. The original segmentation network is augmented to handle this secondary correction task. Both tasks build upon the core feature extraction layers of the model. For the correction task, features detected in the input image are fused with features detected in the predicted segmentation and further processed with task-specific decoder layers. The combination of image and segmentation features allows the model to correct present mistakes in the given input pair. The correction task is trained jointly on the labeled data. On unlabeled data, the exponential moving average of the original network corrects the student’s prediction. The combined outputs of the students’ prediction with the teachers’ correction form the basis for the semi-supervised update. We evaluate our method with the 2017 and 2018 Robotic Scene Segmentation data, the ISIC 2017 and the BraTS 2020 Challenges, a proprietary Endoscopic Submucosal Dissection dataset, Cityscapes, and Pascal VOC 2012. Additionally, we analyze the impact of the individual components and examine the behavior when the amount of labeled data varies, with experiments performed on two distinct segmentation architectures. Our method shows improvements in terms of the mean Intersection over Union over the supervised baseline and competing methods. Code is available at https://github.com/CloneRob/ECMT.
Background and Objectives
Older adults in long-term residential aged care experience loneliness and reduced quality of life (QoL). Biographical approaches use the recall of the past events to increase self-efficacy, promote acceptance, and improve QoL. The aim of the systematic review was to examine the effectiveness of biographical approaches in improving the QoL of older adults in long-term residential aged care.
Research Design and Methods
This systematic review was conducted in accordance with PRISMA guidelines, using databases Medline, CINAHL, and the Cochrane Library. A 2-phase search strategy was used to identify research literature relating to the use of biographical interventions. Narrative analysis was used to synthesize results.
Results
Twenty-one studies met inclusion criteria for this review, comprising both group and individual interventions. The findings were inconsistent. Subjective elements of QoL of older adults improved in a narrow majority of the studies. Of a total of 21 outcomes investigated in group interventions, 11 resulted in significant improvements in QoL. The 16 outcomes of the individual interventions resulted in 10 significant improvements. The most significant influence was observed in life satisfaction. Group reminiscence also had a particularly strong influence on self-esteem.
Discussion and Implications
Although benefits were observed, it remains unclear why some interventions led to improvements in subjective elements of QoL and others did not. Exploration of the participant perspective through qualitative data collection and more detailed description of interventions in future studies could lead to a better understanding of the treatment components that are related to improved outcomes for older adults.
This study presents a new measure of value systems, the Motivational Value Systems Questionnaire (MVSQ), which is based on a theory of value systems by psychologist Clare W. Graves. The purpose of the instrument is to help people identify their personal hierarchies of value systems and thus become more aware of what motivates and demotivates them in work-related contexts. The MVSQ is a forced-choice (FC) measure, making it quicker to complete and more difficult to intentionally distort, but also more difficult to assess its psychometric properties due to ipsativity of FC data compared to rating scales. To overcome limitations of ipsative data, a Thurstonian IRT (TIRT) model was fitted to the questionnaire data, based on a broad sample of <i>N</i> = 1,217 professionals and students. Comparison of normative (IRT) scale scores and ipsative scores suggested that MVSQ IRT scores are largely freed from restrictions due to ipsativity and thus allow interindividual comparison of scale scores. Empirical reliability was estimated using a sample-based simulation approach which showed acceptable and good estimates and, on average, slightly higher test-retest reliabilities. Further, validation studies provided evidence on both construct validity and criterion-related validity. Scale score correlations and associations of scores with both age and gender were largely in line with theoretically - and empirically-based expectations, and results of a multitrait-multimethod analysis supports convergent and discriminant construct validity. Criterion validity was assessed by examining the relation of value system preferences to departmental affiliation which revealed significant relations in line with prior hypothesizing. These findings demonstrate the good psychometric properties of the MVSQ and support its application in the assessment of value systems in work-related contexts.
The success of artificial intelligence in medicine is based on the need for large amounts of high quality training data. Sharing of medical image data, however, is often restricted by laws such as doctor-patient confidentiality. Although there are publicly available medical datasets, their quality and quantity are often low. Moreover, datasets are often imbalanced and only represent a fraction of the images generated in hospitals or clinics and can thus usually only be used as training data for specific problems. The introduction of generative adversarial networks (GANs) provides a mean to generate artificial images by training two convolutional networks. This paper proposes a method which uses GANs trained on medical images in order to generate a large number of artificial images that could be used to train other artificial intelligence algorithms. This work is a first step towards alleviating data privacy concerns and being able to publicly share data that still contains a substantial amount of the information in the original private data. The method has been evaluated on several public datasets and quantitative and qualitative tests showing promising results.
At risk of reproductive disadvantage? Exploring fertility awareness among migrant women in Germany
(2022)
This study examined awareness about fertility among immigrant women and non-migrants in Germany. The social relevance of infertility and fertility treatment is increasing in Western countries due to continually low overall birth rates, a high rate of childlessness, and a gap between the desired and actual numbers of children. While there is growing interest in infertility and reproductive medicine in general, previous studies have rarely included immigrant or ethnic minorities in Europe. This study investigated whether knowledge on the age-related fertility decline (ARFD) varies between migrant groups and the majority group, and the role of education. Working hypotheses were drawn from theoretical considerations on frameworks of migrant assimilation. The analysis was based on data collected in a social science pilot study on reproductive medicine, representative of the general population (‘NeWiRe’ 2014–2015). The sample included 962 women aged 18–50 years living in Germany. Approximately 81% of the sample were immigrants who originated from Turkey, Poland, the Balkan countries or countries of the (post-Soviet) Commonwealth of Independent States. While rather poor overall, knowledge on ARFD was found to be significantly lower in the migrant groups compared with the majority group. This minority-group disadvantage cannot be explained by sociodemographic or cultural variables. Future research should include minority groups in empirical studies on awareness about fertility in order to better understand the causes of this disadvantage, and the potential reproductive needs of migrants.
Physical inactivity is a problem in Europe, contributing to various non-communicable diseases (NCDs). While health sciences offer data and models for preventing the development of NCDs through health promotion, they do not explain the dynamics between the different institutions and actors in the health field. Neofunctionalism refers to these different actors and suggests that their interaction leads to the construction of a supranational authority, allowing the actors to cooperate on common policies, rules and institutions, resulting in integration and Europeanization. Indeed, there are a variety of European Union (EU) institutions and policies concerned with physical activity (PA), however, the purpose of this paper is to analyze if Europeanization and integration are happening in PA promotion, through the lens of neofunctionalism. Analysis uses process tracing for investigating the emergence of PA on the EU agenda, the legal background, existing policies and institutions. Findings indicate that Europeanization and integration of PA promotion are happening, although the opportunity for actions of the EU are restricted to voluntary actions of the EU Member States due to the agreements in the Treaty of Lisbon. This only allows the EU institutions to apply soft law approaches on PA promotion, which seems to be implemented in some Member States despite its non-binding character. The findings are discussed and further implications for public health professionals and policies outlined.