Not seeing the forest for the trees: combination of path integration and landmark cues in human virtual navigation.

Introduction: In order to successfully move from place to place, our brain often combines sensory inputs from various sources by dynamically weighting spatial cues according to their reliability and relevance for a given task. Two of the most important cues in navigation are the spatial arrangement of landmarks in the environment, and the continuous path integration of travelled distances and changes in direction. Several studies have shown that Bayesian integration of cues provides a good explanation for navigation in environments dominated by small numbers of easily identifiable landmarks. However, it remains largely unclear how cues are combined in more complex environments. Methods: To investigate how humans process and combine landmarks and path integration in complex environments, we conducted a series of triangle completion experiments in virtual reality, in which we varied the number of landmarks from an open steppe to a dense forest, thus going beyond the spatially simple environments that have been studied in the past. We analysed spatial behaviour at both the population and individual level with linear regression models and developed a computational model, based on maximum likelihood estimation (MLE), to infer the underlying combination of cues. Results: Overall homing performance was optimal in an environment containing three landmarks arranged around the goal location. With more than three landmarks, individual differences between participants in the use of cues are striking. For some, the addition of landmarks does not worsen their performance, whereas for others it seems to impair their use of landmark information. Discussion: It appears that navigation success in complex environments depends on the ability to identify the correct clearing around the goal location, suggesting that some participants may not be able to see the forest for the trees.

The Virtual Navigation Toolbox: Providing tools for virtual navigation experiments.

Spatial navigation research in humans increasingly relies on experiments using virtual reality (VR) tools, which allow for the creation of highly flexible, and immersive study environments, that can react to participant interaction in real time. Despite the popularity of VR, tools simplifying the creation and data management of such experiments are rare and often restricted to a specific scope-limiting usability and comparability. To overcome those limitations, we introduce the Virtual Navigation Toolbox (VNT), a collection of interchangeable and independent tools for the development of spatial navigation VR experiments using the popular Unity game engine. The VNT's features are packaged in loosely coupled and reusable modules, facilitating convenient implementation of diverse experimental designs. Here, we depict how the VNT fulfils feature requirements of different VR environments and experiments, guiding through the implementation and execution of a showcase study using the toolbox. The presented showcase study reveals that homing performance in a classic triangle completion task is invariant to translation velocity of the participant's avatar, but highly sensitive to the number of landmarks. The VNT is freely available under a creative commons license, and we invite researchers to contribute, extending and improving tools using the provided repository.

Deep-learning based detection of vessel occlusions on CT-angiography in patients with suspected acute ischemic stroke.

Swift diagnosis and treatment play a decisive role in the clinical outcome of patients with acute ischemic stroke (AIS), and computer-aided diagnosis (CAD) systems can accelerate the underlying diagnostic processes. Here, we developed an artificial neural network (ANN) which allows automated detection of abnormal vessel findings without any a-priori restrictions and in <2 minutes. Pseudo-prospective external validation was performed in consecutive patients with suspected AIS from 4 different hospitals during a 6-month timeframe and demonstrated high sensitivity (≥87%) and negative predictive value (≥93%). Benchmarking against two CE- and FDA-approved software solutions showed significantly higher performance for our ANN with improvements of 25-45% for sensitivity and 4-11% for NPV (p ≤ 0.003 each). We provide an imaging platform ( https://stroke.neuroAI-HD.org ) for online processing of medical imaging data with the developed ANN, including provisions for data crowdsourcing, which will allow continuous refinements and serve as a blueprint to build robust and generalizable AI algorithms.

AI-Supported Comprehensive Detection and Quantification of Biomarkers of Subclinical Widespread Diseases at Chest CT for Preventive Medicine.

Automated image analysis plays an increasing role in radiology in detecting and quantifying image features outside of the perception of human eyes. Common AI-based approaches address a single medical problem, although patients often present with multiple interacting, frequently subclinical medical conditions. A holistic imaging diagnostics tool based on artificial intelligence (AI) has the potential of providing an overview of multi-system comorbidities within a single workflow. An interdisciplinary, multicentric team of medical experts and computer scientists designed a pipeline, comprising AI-based tools for the automated detection, quantification and characterization of the most common pulmonary, metabolic, cardiovascular and musculoskeletal comorbidities in chest computed tomography (CT). To provide a comprehensive evaluation of each patient, a multidimensional workflow was established with algorithms operating synchronously on a decentralized Joined Imaging Platform (JIP). The results of each patient are transferred to a dedicated database and summarized as a structured report with reference to available reference values and annotated sample images of detected pathologies. Hence, this tool allows for the comprehensive, large-scale analysis of imaging-biomarkers of comorbidities in chest CT, first in science and then in clinical routine. Moreover, this tool accommodates the quantitative analysis and classification of each pathology, providing integral diagnostic and prognostic value, and subsequently leading to improved preventive patient care and further possibilities for future studies.

Patterns of Safety Incidents in a Neonatal Intensive Care Unit.

Introduction: Safety incidents preceding manifest adverse events are barely evaluated in neonatal intensive care units (NICUs). This study aimed at identifying frequency and patterns of safety incidents in our NICU. Methods: A 6-month prospective clinical study was performed from May to October 2019 in a German 10-bed level III NICU. A voluntary, anonymous reporting system was introduced, and all neonatal team members were invited to complete paper-based questionnaires following each particular safety incident. Safety incidents were defined as safety-related events that were considered by the reporting team member as a "threat to the patient's well-being" which "should ideally not occur again." Results: In total, 198 safety incidents were analyzed. With 179 patients admitted, the incident/admission ratio was 1.11. Medication errors (n = 94, 47%) and equipment problems (n = 54, 27%) were most commonly reported. Diagnostic errors (n = 19, 10%), communication problems (n = 12, 6%), errors in documentation (n = 9, 5%) and hygiene problems (n = 10, 5%) were less frequent. Most safety incidents were noticed after 4-12 (n = 52, 26%) and 12-24 h (n = 47, 24%), respectively. Actual harm to the patient was reported in 17 cases (9%) but no life-threatening or serious events occurred. Of all safety incidents, 184 (93%) were considered to have been preventable or likely preventable. Suggestions for improvement were made in 132 cases (67%). Most often, implementation of computer-assisted tools and processes were proposed. Conclusion: This study confirms the occurrence of various safety incidents in the NICU. To improve quality of care, a graduated approach tailored to the specific problems appears to be prudent.

Joint Imaging Platform for Federated Clinical Data Analytics.

PURPOSE: Image analysis is one of the most promising applications of artificial intelligence (AI) in health care, potentially improving prediction, diagnosis, and treatment of diseases. Although scientific advances in this area critically depend on the accessibility of large-volume and high-quality data, sharing data between institutions faces various ethical and legal constraints as well as organizational and technical obstacles. METHODS: The Joint Imaging Platform (JIP) of the German Cancer Consortium (DKTK) addresses these issues by providing federated data analysis technology in a secure and compliant way. Using the JIP, medical image data remain in the originator institutions, but analysis and AI algorithms are shared and jointly used. Common standards and interfaces to local systems ensure permanent data sovereignty of participating institutions. RESULTS: The JIP is established in the radiology and nuclear medicine departments of 10 university hospitals in Germany (DKTK partner sites). In multiple complementary use cases, we show that the platform fulfills all relevant requirements to serve as a foundation for multicenter medical imaging trials and research on large cohorts, including the harmonization and integration of data, interactive analysis, automatic analysis, federated machine learning, and extensibility and maintenance processes, which are elementary for the sustainability of such a platform. CONCLUSION: The results demonstrate the feasibility of using the JIP as a federated data analytics platform in heterogeneous clinical information technology and software landscapes, solving an important bottleneck for the application of AI to large-scale clinical imaging data.

[Results of the hypothermia registry of the University Hospital Dresden from 2010-2017. Application of hypothermia therapy in newborns in Germany]. / Ergebnisse des Hypothermienetzwerks des Universitätsklinikums Dresden von 2010­2017. Entwicklung der Hypothermietherapie bei Neugeborenen in Deutschland.

INTRODUCTION: Therapeutic hypothermia (TH) improves the outcome in newborns with hypoxic-ischemic encephalopathy (HIE) and should be used in case of perinatal asphyxia and signs of moderate/severe HIE. MATERIAL/METHODS: Frequency of HIE and the application of TH were extracted from the neonatal survey, a registry that collects data from all German hospitals, and from the hypothermia registry, established in 2010. The latter was also used to analyze short-term outcomes of the newborns. RESULTS: Between 2010 and 2017, 106 of Germany's 213 perinatal centers joined the registry. Response rates varied between 22 and 60%. The registry recorded 164 (IQR 115-224) TH cases per year in newborns with HIE. In the neonatal survey, 517 (382-664) TH and 543 (432-581) HIE cases were reported. Since 2014 there have been more cases of TH than HIE. After TH, 10.4% (8-13%) of the newborns died, 81% (78-82%) of the newborns were discharged home, 3.6% (3-5%) to a rehabilitation facility, and 5.4% (5-7%) transferred to another clinic. 89% (87-89%) were on complete oral feedings. DISCUSSION: After the introduction of TH in the clinical routine, the number of treated newborns increased continuously. Currently, the number of TH is higher than the number of children with HIE, which is difficult to explain, as the presence of a moderate or severe HIE is a mandatory requirement for TH. The data from the hypothermia registry showed no significant changes in mortality or neurological outcome over time.

The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-based Phenotyping.

Background Radiomic features may quantify characteristics present in medical imaging. However, the lack of standardized definitions and validated reference values have hampered clinical use. Purpose To standardize a set of 174 radiomic features. Materials and Methods Radiomic features were assessed in three phases. In phase I, 487 features were derived from the basic set of 174 features. Twenty-five research teams with unique radiomics software implementations computed feature values directly from a digital phantom, without any additional image processing. In phase II, 15 teams computed values for 1347 derived features using a CT image of a patient with lung cancer and predefined image processing configurations. In both phases, consensus among the teams on the validity of tentative reference values was measured through the frequency of the modal value and classified as follows: less than three matches, weak; three to five matches, moderate; six to nine matches, strong; 10 or more matches, very strong. In the final phase (phase III), a public data set of multimodality images (CT, fluorine 18 fluorodeoxyglucose PET, and T1-weighted MRI) from 51 patients with soft-tissue sarcoma was used to prospectively assess reproducibility of standardized features. Results Consensus on reference values was initially weak for 232 of 302 features (76.8%) at phase I and 703 of 1075 features (65.4%) at phase II. At the final iteration, weak consensus remained for only two of 487 features (0.4%) at phase I and 19 of 1347 features (1.4%) at phase II. Strong or better consensus was achieved for 463 of 487 features (95.1%) at phase I and 1220 of 1347 features (90.6%) at phase II. Overall, 169 of 174 features were standardized in the first two phases. In the final validation phase (phase III), most of the 169 standardized features could be excellently reproduced (166 with CT; 164 with PET; and 164 with MRI). Conclusion A set of 169 radiomics features was standardized, which enabled verification and calibration of different radiomics software. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Kuhl and Truhn in this issue.