Search

Fracture detection by artificial intelligence and especially Deep Convolutional Neural Networks (DCNN) is a topic of growing interest in current orthopaedic and radiological research. As learning a DCNN usually needs a large amount of training data, mostly frequent fractures as well as conventional X-ray are used. Therefore, less common fractures like acetabular fractures (AF) are underrepresented in the literature. The aim of this pilot study was to establish a DCNN for detection of AF using computer tomography (CT) scans.Patients with an acetabular fracture were identified from the monocentric consecutive pelvic injury registry at the BG Trauma Center XXX from 01/2003 - 12/2019. All patients with unilateral AF and CT scans available in DICOM-format were included for further processing. All datasets were automatically anonymised and digitally post-processed. Extraction of the relevant region of interests was performed and the technique of data augmentation (DA) was implemented to artificially increase the number of training samples. A DCNN based on Med3D was used for autonomous fracture detection, using global average pooling (GAP) to reduce overfitting.From a total of 2,340 patients with a pelvic fracture, 654 patients suffered from an AF. After screening and post-processing of the datasets, a total of 159 datasets were enrolled for training of the algorithm. A random assignment into training datasets (80%) and test datasets (20%) was performed. The technique of bone area extraction, DA and GAP increased the accuracy of fracture detection from 58.8% (native DCNN) up to an accuracy of 82.8% despite the low number of datasets.The accuracy of fracture detection of our trained DCNN is comparable to published values despite the low number of training datasets. The techniques of bone extraction, DA and GAP are useful for increasing the detection rates of rare fractures by a DCNN. Based on the used DCNN in combination with the described techniques from this pilot study, the possibility of an automatic fracture classification of AF is under investigation in a multicentre study.Die automatische Frakturerkennung mithilfe von künstlicher Intelligenz (KI) ist derzeit ein wichtiges Thema in der unfallchirurgischen und radiologischen Forschung. Vor allem bei der Nutzung von „Deep Convolutional Neural Networks“ (DCNN) ist allerdings üblicherweise eine große Anzahl an Trainingsdatensätzen erforderlich, um gute Erkennungsraten zu erzielen, weswegen diese Methoden bisher i. d. R. an sehr häufigen Frakturen und an Röntgenbildern angewendet werden. Demgegenüber stellen seltene Frakturen wie die Azetabulumfraktur (AF) eine Herausforderung für neuronale Netzwerke dar. Ziel dieser Pilotstudie war es daher, einen Algorithmus zu entwickeln, mit dessen Hilfe AF anhand von CT-Bildern zuverlässig automatisch erkannt werden können.Patienten mit einer einseitigen AF wurden aus dem monozentrischen Beckenregister der BG Unfallklinik Tübingen über einen Zeitraum von 01/2003 bis 12/2019 identifiziert. Patienten, bei denen CT-Datensätze im DICOM-Format vorlagen, wurden weiterverarbeitet. Die automatische Weiterverarbeitung beinhaltete die Anonymisierung sowie verschiedene Methoden der Knochenextraktion, der „Data Augmentation“ (DA) sowie des Global Average Poolings (GAP), um die Datensätze für das Training des auf Basis von Med3D erstellten DCNN vorzubereiten.Von insgesamt 2340 Patienten mit einer Beckenfraktur erlitten 654 eine AF. Nach Weiterverarbeitung der Datensätze konnten 159 Datensätze für das Training verwendet werden. 80% davon wurden als Trainingsdatensätze randomisiert, 20% wurden als Testdatensätze randomisiert. Durch die Techniken der DA und des GAP konnte die Erkennungsrate des DCNN trotz der geringen Zahl an Trainingsdatensätzen von 58,8 auf 82,8% gesteigert werden.Die Erkennungsrate unseres DCNN ist trotz der im Vergleich niedrigen Zahl an Trainingsdatensätzen vergleichbar mit Literaturangaben bei häufigeren Frakturen. Die Techniken der Knochenextraktion, verbunden mit der DA und der GAP, helfen bei selten auftretenden Frakturen, die Erkennungsraten eines DCNN zu erhöhen. Mit den Techniken und dem DCNN aus dieser Pilotstudie soll in einer Multicenterstudie an einer größeren Anzahl an Datensätzen die Möglichkeit einer automatischen Klassifizierung von AF untersucht werden.

In many surgical disciplines, the interest among medical students to pursue a surgical career decreases during their medical studies. The same goes for students after graduation. The aim of our cohort study was to evaluate the operating room (OR) experiences of medical students during our curriculum. Over the course of one year 217 senior year medical students were included in our study. All of them took part in our training program for senior year medical students, which consisted of a 1-week clinical rotation including visits to the OR. We developed a Likert-scaled questionnaire, which was evaluated anonymously; free text answers were also possible. Prior to the analysis of the sex and age differences, we confirmed that the evaluation scale provided a coherent measure of the OR evaluation. As a first proxy, we conducted a series of Spearman correlations which revealed high intercorrelations between all of the six items of the questionnaire, r(154) = 0.53 to r(154) = 0.94, all p

The study examines whether the recognition and attribution of emotions with regard to a specific situation is related to own experiences of a similar situation or conceivable experiences of a similar situation in near future.

Bias in clinical reasoning has been identified as one of the major sources of diagnostic errors (Croskerry, 2013). While there have been several studies investigating the effectiveness of different, mostly cognitive, strategies to counteract or mitigate bias (Mamede et al., 2010, 2020; Van Den Berge & Mamede, 2013), to our knowledge empirical evidence on successfully counteracting such biases by utilizing new possibilities provided by technology is lacking. A study by Crowley et al. (2013) developed an automated, computer-based method of recognizing heuristics and biases, showing that the detection of biases by technology can be possible. Wang et al. (2019) also recognize the new possibilities of artificial intelligence (AI) in potentially counteracting bias, offering strategies to mitigate various biases. In this study we want to investigate the efficacy of presenting additional information through a simulated decision support system (DSS) to debias medical personal. There are more than 100 failed heuristics or cognitive biases in the medical context, as listed by Croskerry (2002). In this study we will focus on two of the most commonly described biases, availability and representativeness (Lucchiari & Pravettoni, 2012), to profit from and build on previous research. Availability bias is “the tendency for things to be judged more frequently if they come readily to mind” (Croskerry, 2002). This includes emotionally salient, more recent (Richie & Josephson, 2018), unusual or adverse events that are more memorable (Wang et al., 2019). The suggested mitigation strategy is gathering of objective information and making an estimation on the true base rate of a diagnosis. According to Wang et al. (2019) support systems might achieve this by showing prior probability of a diagnosis to help seek the base rate. Representativeness heuristics and the subsequent bias entails judging the likelihood of an event belonging to a particular class due to similarities and thereby disregarding whether it belongs to a similar but different class of diagnoses (Wang et al., 2019). Wang et al. (2019) suggest mitigating this by comparing the disease with prototypes of the conditions and identifying similarity distances as well as contrasting differences. Additionally, we would like to look into some demographic factors of participants to investigate potential influence on the probability to exhibit bias and the tendency to change their estimate when consulted by a DSS. Therefore, age, gender and medical experience will be measured. Age, gender and experience are moderating variables in the extended technology acceptance model designed by Venkatesh et al. (2003), however, Shibl et al. (2013) found no significant influence of age and gender on the acceptance and therefore the usage of clinical DSS. If medical personal is not using the DSS or accepting the consultation, they cannot profit from the support. Therefore, we posit the research question: Does gender and age have an influence on the probability to adapt one’s estimate to received decision aid? Furthermore, the higher the confidence in the initial diagnosis the less likely it is for the physician to adapt their diagnosis (Dreiseitl & Binder, 2005). On the same note, higher experience of the physicians was shown to correlate negatively with the likelihood of changing the diagnosis (Dreiseitl & Binder, 2005). Therefore, we want to investigate the influence of confidence and experience on the probability to adapt the estimated probability of diagnosis according to support from a DSS. Additionally, Wang et al. (2019) mentioned, that physicians explained that they preferred to form a first uninfluenced opinion before receiving support from a DSS. Meanwhile Kostopoulou et al. (2015) found that reminding general practitioners early of possible diagnoses lead to higher accuracy. Hence, we want to investigate, whether there is a difference between the group immediately receiving decision aid and the group giving a first uninfluenced estimation. Therefore we posit the following research question: Does the time of decision aid have an influence on the final estimate? In this study, we want to investigate, whether a simulated DSS (based on the abilities of existing and discussed AI) is able to mitigate the effect of availability and representativeness bias on physician’s diagnosis. Therefore, we aim to achieve the following goals: 1) To test debiasing methods that were proposed by previous research and find out whether they can succeed in mitigating the biases. 2) To investigate further influencing factors like confidence and experience on the probability to elicit bias and remedy the decision after receiving decision aid. 3) To investigate whether the time of support and the possibility to form a first opinion without decision aid makes a difference.

This thesis aims to evaluate the effects of a face-to-face online intergroup dialogue program (Soliya Connect Global) including participants originating from various social and cultural identity groups and different continents. The focus will be to investigate how the participation influences intergroup attitudes, such as prejudices (feeling towards different groups), outgroup overlap, perspective taking and behavioural measures, e.g., challenging media representations, promoting awareness and communicating interculturally, in comparison with an untrained control group.

Physical training has beneficial effects not only on physical fitness, but also on cognitive functions. The most effective way to improve cognitive functions via physical training as well as the degree to which training effects transfer to untrained cognitive functions is still unclear, however. Here, we investigated the effects of adaptive and multi-modal short-term training interventions on cognitive training gains and transfer effects. Over a period of 12 weeks, 102 employees of a car manufacturing company (age range 20 to 61 years) received trainer-guided exercises, consisting of either two adaptive training interventions, physical (strength) training and multi-modal (motor–cognitive) training, or non-adaptive strength training (active control group). For the multi-modal intervention, the “Agility Board” was employed, a novel, multi-modal training device. Pre- and post-training, psychometric tests were conducted to measure cognitive abilities, such as perceptual speed, attention, short-term memory, working memory, inhibition, and mental rotation. In addition, motor–cognitive performance was assessed. Compared with the active control group, both training groups showed enhanced performance at posttest. While multi-modal training yielded performance improvements only in trained tasks, physical training was associated with improvements in untrained working memory updating and immediate recall tasks, suggesting transfer effects to short-term and working memory functioning. In summary, the results demonstrate the importance of adaptive difficulty settings for short-term physical training interventions, at least for the enhancement of working memory.

Hochschulische Kompetenzen von Studienabbrecher:innen können bei der Aufnahme in einigen Fachschulbildungsgängen in NRW pauschal angerechnet werden. Die mit einer Anrechnung einhergehende Verkürzung der Weiterbildungsdauer ermöglicht eine schnellere Integration von Studierenden in den Arbeitsmarkt.

Analgesic consumption is highly prevalent in Germany. About 1.9 million people consume analgesics daily; around 1.6 million engage in addictive use. Analgesic consumption also seems to be widespread in sports. We aim to quantify the degree of analgesic consumption in sports on the basis of a systematic literature review.

The dynamics of the size distribution in growing and dividing rod-shaped bacteria can change depending on the perspective through which the cell population is observed. From the perspective of single-cell lineages, only one descendant is followed after each division, while in population snapshots all of the offspring is considered. In this letter, we propose an analytical derivation of Population Balance Equations (PBEs) that can be used to compute the dynamics of the main statistics of the cell size distribution. We then explore how the dynamics are affected by different division strategies and by the two different perspectives of observation. To assess the accuracy of our derivations, we compare the analytical results with those obtained numerically and through stochastic simulations. The comparison allows us to establish mathematical relationships between the two perspectives of observation, thus improving our knowledge of how cells control their growth and proliferation.

Phenotypic variability in cell populations can be described through population balance models (PBM). In this paper, the simulation of cell PBMs via two different methods is compared. We compare the cell average technique (CAT) with the recently introduced numerical Gaussian processes (GPs). While the CAT discretises the PBM in the individual variables, i.e. in the cellular properties, numerical GPs circumvent this discretisation. Numerical GPs are a probabilistic machine-learning method that additionally provides uncertainty quantification. Using two simulation case studies, we compare both methods with regards to computational load, accuracy, and implementation effort. In one of the two considered case studies, numerical GPs are inferior to the CAT. Nonetheless, they are an attractive option when uncertainty quantification is very much required, such as in a subsequent state estimation for PBMs.

Past research has neglected the question of whether different exchange mechanisms and consumption or service contexts (CSCs) in the sharing economy may lead to different outcomes. The authors develop a novel theoretical framework that proposes that willingness to participate in the sharing economy will be lower if exchange mechanisms and CSCs do not match—that is, if monetary/social exchange mechanisms are applied in hedonic/utilitarian CSCs, taking into account the moderating role of materialism. Moreover, the authors propose that in the sharing economy, monetary exchanges may cause a crowding-out of prosocial behaviors and that this effect will be especially pronounced for highly materialistic individuals. They test their hypothesized framework in three experimental studies that support the theorizing and provide results with important implications for the design of sharing economy offers as well as for society.

Physical training has beneficial effects not only on physical fitness, but also on cognitive functions. The most effective way to improve cognitive functions via physical training as well as the degree to which training effects transfer to untrained cognitive functions is still unclear, however. Here, we investigated the effects of adaptive and multi-modal short-term training interventions on cognitive training gains and transfer effects. Over a period of 12 weeks, 102 employees of a car manufacturing company (age range 20 to 61 years) received trainer-guided exercises, consisting of either two adaptive training interventions, physical (strength) training and multi-modal (motor-cognitive) training, or non-adaptive strength training (active control group). For the multi-modal intervention, the "Agility Board" was employed, a novel, multi-modal training device. Pre- and post-training, psychometric tests were conducted to measure cognitive abilities, such as perceptual speed, attention, short-term memory, working memory, inhibition, and mental rotation. In addition, motor-cognitive performance was assessed. Compared with the active control group, both training groups showed enhanced performance at posttest. While multi-modal training yielded performance improvements only in trained tasks, physical training was associated with improvements in untrained working memory updating and immediate recall tasks, suggesting transfer effects to short-term and working memory functioning. In summary, the results demonstrate the importance of adaptive difficulty settings for short-term physical training interventions, at least for the enhancement of working memory.

Cement augmentation of sacroiliac (SI) screws in the posterior pelvic ring has been shown to provide greater biomechanical stability in cadaveric studies. Pelvic ring fractures are relatively rare compared to the total number of fractures. Nevertheless, the 1‑year mortality rate of up to 27% is very high, especially for geriatric pelvic ring fractures and is also largely associated with reduced mobility due to the fracture. The primary goal of treatment is therefore the restoration of patient mobility. This requires the achievement of sufficient stability of the pelvic ring. As osteoporosis is often a causative factor for the pelvic ring fracture, a more stable anchoring of the implants in the osteoporotic bone can be achieved by cement augmentation. This article presents the possibilities of cement augmentation of the pelvic ring and describes the technique of cement-augmented SI screws.Die Zementaugmentation von transiliosakralen (SI) Schrauben am hinteren Beckenring sorgt in Kadaverstudien nachgewiesenermaßen für eine höhere biomechanische Stabilität. Beckenringfrakturen sind gemessen an der Gesamtzahl der Frakturen eher selten. Dennoch ist die 1‑Jahres-Mortalität insbesondere bei geriatrischen Beckenringfrakturen mit bis zu 27 % sehr hoch und zu einem Großteil auch auf die reduzierte Mobilität aufgrund der Fraktur zurückzuführen. Oberstes Ziel der Behandlung ist die Wiederherstellung der Mobilität der Patientinnen und Patienten. Hierfür ist das Erreichen einer ausreichenden Stabilität des Beckenrings erforderlich. Da oftmals eine Osteoporose mitursächlich für die Beckenringfraktur ist, kann durch eine Zementaugmentation eine stabilere Verankerung der Implantate im osteoporotischen Knochen erreicht werden. In dieser Arbeit werden die Möglichkeiten der Zementaugmentation am Beckenring aufgezeigt und die Technik der zementaugmentierten SI-Schraube beschrieben.

Neoadjuvant chemoradiotherapy (CRT) followed by surgery is recommended for patients with diagnosed rectal cancer UICC stage II/III. The present study aimed to evaluate the accuracy of preoperative staging with focus on tumor infiltration depth and lymph node status challenging the indication of neoadjuvant CRT. All consecutive rectal cancer patients who underwent surgical resection without neoadjuvant CRT at the Klinikum Stuttgart, Germany, between January 2015 and December 2018, were included into the study. Clinicopathologic features focusing on preoperative tumor staging and histological outcome were assessed. A total of 100/162 patients (61.7%) underwent primary surgical rectal resection with curative intent. Among these patients, 54/100 had a correct preoperative T-staging, while 34 were overstaged and 12 understaged. With regard to the nodal status, 68 were accurately staged, while 28 were overstaged and 4 understaged. Only 4/40 perirectal lymph nodes of more than 5 mm in diameter in preoperative MRI histologically revealed to be metastasis. For patients without neoadjuvant CRT, a tendency to preoperative overstaging was observed. Lymph node size alone did not reliably predict metastasis. According to current guidelines, 21/62 (33.9%) of these patients would have been overtreated by using CRT. On the background of relevant side effects, complications, and the limited benefit of CRT on overall survival, we suggest that primary surgical resection should be recommended more liberally for stages II and III rectal cancer.

Zusammenfassung Reflexion gilt im Lehrer*innenbildungsdiskurs als Qualitätsgarant. Mit ihr soll das Problem routinisierten Denkens und Handelns bearbeitet werden, indem gewohnte Denk- und Handlungsmuster unter Bezug auf wissenschaftliches Wissen hinterfragt werden. Empirische Untersuchungen zeigen aber, dass Reflexion dieser Art in schulpraktischen Reflexionsformaten kaum stattfindet. Diese Beobachtung wird in dem Beitrag zum Anlass genommen, um ausgehend von einem weiten Verständnis von Reflexion als Rückbeuge auf die eigene Praxis zu untersuchen, wie sich Reflexion in der praktischen Lehrer*innenbildung vollzieht, um von hier aus den Reflexionsbegriff des Lehrer*innenbildungsdiskurses weiterzuentwickeln. Hierfür wird eine Szene aus einem Unterrichtsnachgespräch, das im Rahmen eines Unterrichtsbesuches im Kontext des Referendariats stattfand, gesprächsanalytisch ausgewertet. Anschließend wird vorgeschlagen, mit transzendierender und situierender Reflexion zwischen zwei Reflexionsformen zu unterscheiden, die jeweils eigene Leistungen in der Ausbildung von Lehrkräften erbringen. Der Beitrag schließt mit der Darstellung von Implikationen der vorgestellten Befunde für die weitere Diskussion von Reflexion in der Lehrer*innenbildung.

Pelvic-ring-injuries are bony and/or ligamentous disruptions that might lead to pelvic instabilities and asymmetries. For surgical treatment, knowledge of the different possible approaches and stabilization-techniques is essential. From 2003 to 2017, we analyzed all complications in operative and non-operative treatment of pelvic-ring-injuries in our institution. Besides complication rates, we describe the reachable anatomical structures for different surgical approaches to the pelvic-ring focusing on strategies to reduce surgical complications. Of 1848 patients with a pelvic fracture, 1463 patients had a pelvic-ring-fracture. We found an overall complication-rate of 14.4%. For reducing approach-related morbidity, preoperative planning and knowledge of approach-related complications is important.

Abstract. Subjects usually respond faster and more accurate in trials in which the response location corresponds to a task-irrelevant stimulus location compared to when not. Previous research has shown, that this so-called Simon effect is reduced after non-corresponding compared to after corresponding trials. As of now it is yet unclear what exact mechanisms drive such sequential modulations. One theory assumes a conflict adaptation mechanism that decreases the influence of incongruent information after non-corresponding trials via increased cognitive control. However, other authors claim that feature integration processes may be the underlying mechanism as the amount of feature overlap differs between different correspondence sequences. Unfortunately, this also means that in the standard Simon task the repetition of task features and correspondence sequences are not independent. In order to address this issue, we mapped four stimuli to two responses in the present EEG study. This way, we created a task, which allows for sequences in which the stimulus identity may change without alternating the required response. These sequences may either comprise a change of the stimulus position or not and the contribution of feature integration as well as conflict adaptation processes could thus be observed independently. Our results indicate that the repetition of task features affects performance to a stronger degree than the correspondence sequence and feature integration processes do. Yet, an impact of both could still be observed. The strongest effect of feature repetitions on task performance was observed for task-relevant features, especially for the imperative stimulus identity itself. The EEG results support these findings. The amplitudes of the fronto-central N2 and the parietal P3 decreased with increasing feature overlap from one trial to the next. The posterior lateralization, reflected by the posterior contralateral negativity (PCN), however, appears to reflect mainly changes in the stimulus location and stimulus–response (S–R) correspondence rather than feature repetitions per se.