Search

Training classifier models with semi-labeled datasets, which often have only a limited number of labeled samples, is challenging. This thesis proposes a user-centric methodology for pseudo-labeling semi-labeled data, fusing automatic pseudo-labeling algorithms with user-driven correction of mislabeled data points. The methodology is supported by a number of visual analytics approaches involving sample visualization via dimensionality reduction techniques and visualization of classifier decision boundaries using so-called Decision Boundary Maps (DBMs). These visuals allow users to find regions of uncertainty where automatic pseudo-labeling may have made errors and correct these accordingly. To speed up the visual analytics loop, we propose various heuristics for efficient and accurate DBM computation. Conducted user experiments show that both domain expert and non-expert users were able to consistently correct wrong labels and improve classifier performance for different datasets and classifier models, with only a limited effort in a limited amount of time. The study underscores the importance and potential of visualization tools in the context of semi-labeled datasets and semi-supervised learning and provides a foundation for future research in this area.

AIM The optimal management for early recurrent prostate cancer following radical prostatectomy (RP) in patients with negative prostate-specific membrane antigen positron-emission tomography (PSMA-PET) scan is an ongoing subject of debate. The aim of this study was to evaluate the outcome of salvage radiotherapy (SRT) in patients with biochemical recurrence with negative PSMA PET finding. METHODS This retrospective, multicenter (11 centers, 5 countries) analysis included patients who underwent SRT following biochemical recurrence (BR) of PC after RP without evidence of disease on PSMA-PET staging. Biochemical recurrence-free survival (bRFS), metastatic-free survival (MFS) and overall survival (OS) were assessed using Kaplan-Meier method. Multivariable Cox proportional hazards regression assessed predefined predictors of survival outcomes. RESULTS Three hundred patients were included, 253 (84.3%) received SRT to the prostate bed only, 46 (15.3%) additional elective pelvic nodal irradiation, respectively. Only 41 patients (13.7%) received concomitant androgen deprivation therapy (ADT). Median follow-up after SRT was 33 months (IQR: 20-46 months). Three-year bRFS, MFS, and OS following SRT were 73.9%, 87.8%, and 99.1%, respectively. Three-year bRFS was 77.5% and 48.3% for patients with PSA levels before PSMA-PET ≤ 0.5 ng/ml and > 0.5 ng/ml, respectively. Using univariate analysis, the International Society of Urological Pathology (ISUP) grade > 2 (p = 0.006), metastatic pelvic lymph nodes at surgery (p = 0.032), seminal vesicle involvement (p < 0.001), pre-SRT PSA level of > 0.5 ng/ml (p = 0.004), and lack of concomitant ADT (p = 0.023) were significantly associated with worse bRFS. On multivariate Cox proportional hazards, seminal vesicle infiltration (p = 0.007), ISUP score >2 (p = 0.048), and pre SRT PSA level > 0.5 ng/ml (p = 0.013) remained significantly associated with worse bRFS. CONCLUSION Favorable bRFS after SRT in patients with BR and negative PSMA-PET follow

AIM: The current work aimed to investigate the clinical benefit of radiotherapy in patients with metastatic non-small cell lung cancer (NSCLC) developing acquired resistance to immune checkpoint inhibitors. METHOD: We report on a pooled, two-institution, phase II single-arm prospective cohort study. The study included patients with stage IV NSCLC who showed progression of one or more measurable lesions under anti-PD-(L)1 inhibition alone, after initially having achieved at least stable disease. Hypofractionated radiotherapy (hRT) of one to four metastases was performed, while one or more lesions were kept untreated. Following hRT, treatment with immune checkpoint inhibitors was continued unchanged until further evidence of tumor progression or unacceptable toxicity. Primary endpoint of the pooled analysis was progression-free survival (PFS), secondary endpoints included overall survival (OS) and toxicity. RESULTS: A total of 48 patients were enrolled: mean age was 67.1 ± 9.3 years, 50% were male and 72.9% were PD-L1 positive. Immunotherapy was in 95.8% of patients the first or second line therapy at time of enrollment. hRT was performed to one (93.8% of cases) or more lesions (median total dose: 27.5 Gy, median 6.5 Gy/fraction). Forty-five patients (93.8%) were able to continue immunotherapy for a median of 6.2 months following hRT. Median PFS was 4.4 months, with 62.5% disease control at three months and 37.5% at six months. Median OS was 14.9 months. Severe adverse events (grade =2) were reported in 12 cases (25%), of which none were radiotherapy-related and four were immunotherapy-related. Salvage therapy consisted of chemotherapy (48.8%) or repeated irradiation (21.9%). No further tumor treatment was performed in 29.3% of patients. CONCLUSIONS: The current pooled analysis is a prospective evaluation of the role of radiation therapy for metastatic NSCLC in the setting of newly acquired immunotherapy resistance. Hypofractionated radiotherapy can support the outcom

The automated generation of diverse and complex training scenarios has been an important ingredient in many complex learning tasks. Especially in real-world application domains, such as autonomous driving, auto-curriculum generation is considered vital for obtaining robust and general policies. However, crafting traffic scenarios with multiple, heterogeneous agents is typically considered as a tedious and time-consuming task, especially in more complex simulation environments. In our work, we introduce MATS-Gym, a Multi-Agent Traffic Scenario framework to train agents in CARLA, a high-fidelity driving simulator. MATS-Gym is a multi-agent training framework for autonomous driving that uses partial scenario specifications to generate traffic scenarios with variable numbers of agents. This paper unifies various existing approaches to traffic scenario description into a single training framework and demonstrates how it can be integrated with techniques from unsupervised environment design to automate the generation of adaptive auto-curricula. The code is available at https://github.com/AutonomousDrivingExaminer/mats-gym., Comment: 7 Pages, Under Review

We introduce Gated Chemical Units (GCUs), a new type of gated recurrent cells which provide fresh insights into the commonly-used gated recurrent units, and bridge their gap to biologically-plausible neural models. We systematically derive GCUs from Electrical Equivalent Circuits (EECs), a widely adopted ordinary-differential-equations model in neuroscience for biological neurons with both electrical and chemical synapses. We focus on saturated EECs, as they are more stable, and chemical synapses, as they are more expressive. To define GCUs, we introduce a new kind of gate, we call a time gate (TG), in the associated difference-equations model of the EECs. The TG learns for each neuron the optimal time step to be used in a simple Euler integration scheme, and leads to a very efficient gated unit. By observing that the TG corresponds to the forget gate (FG) in traditional gated recurrent units, we provide a new formulation of these units as neural ODEs. We also show that in GCUs, the FG is in fact its liquid time constant. Finally, we demonstrate that GCUs not only explain the elusive nature of gates in traditional recurrent units, but also represent a very competitive alternative to these units.

Recent advances in depthwise-separable convolutional neural networks (DS-CNNs) have led to novel architectures, that surpass the performance of classical CNNs, by a considerable scalability and accuracy margin. This paper reveals another striking property of DS-CNN architectures: discernible and explainable patterns emerge in their trained depthwise convolutional kernels in all layers. Through an extensive analysis of millions of trained filters, with different sizes and from various models, we employed unsupervised clustering with autoencoders, to categorize these filters. Astonishingly, the patterns converged into a few main clusters, each resembling the difference of Gaussian (DoG) functions, and their first and second-order derivatives. Notably, we were able to classify over 95\% and 90\% of the filters from state-of-the-art ConvNextV2 and ConvNeXt models, respectively. This finding is not merely a technological curiosity; it echoes the foundational models neuroscientists have long proposed for the vision systems of mammals. Our results thus deepen our understanding of the emergent properties of trained DS-CNNs and provide a bridge between artificial and biological visual processing systems. More broadly, they pave the way for more interpretable and biologically-inspired neural network designs in the future.

In this study, we present evidence suggesting that depthwise convolutional kernels are effectively replicating the structural intricacies of the biological receptive fields observed in the mammalian retina. We provide analytics of trained kernels from various state-of-the-art models substantiating this evidence. Inspired by this intriguing discovery, we propose an initialization scheme that draws inspiration from the biological receptive fields. Experimental analysis of the ImageNet dataset with multiple CNN architectures featuring depthwise convolutions reveals a marked enhancement in the accuracy of the learned model when initialized with biologically derived weights. This underlies the potential for biologically inspired computational models to further our understanding of vision processing systems and to improve the efficacy of convolutional networks.

Background: The aim of this study was to investigate the role of clinical, dosimetric and pretherapeutic magnetic resonance imaging (MRI) features for lesion-specific outcome prediction of stereotactic radiotherapy (SRT) in patients with brain metastases from malignant melanoma (MBM). Methods: In this multicenter, retrospective analysis, we reviewed 517 MBM from 130 patients treated with SRT (single fraction or hypofractionated). For each gross tumor volume (GTV) 1576 radiomic features (RF) were calculated (788 each for the GTV and for a 3 mm margin around the GTV). Clinical parameters, radiation dose and RF from pretherapeutic contrast-enhanced T1-weighted MRI from different institutions were evaluated with a feature processing and elimination pipeline in a nested cross-validation scheme. Results: Seventy-two (72) of 517 lesions (13.9%) showed a local failure (LF) after SRT. The processing pipeline showed clinical, dosimetric and radiomic features providing information for LF prediction. The most prominent ones were the correlation of the gray level co-occurrence matrix of the margin (hazard ratio (HR): 0.37, confidence interval (CI): 0.23-0.58) and systemic therapy before SRT (HR: 0.55, CI: 0.42-0.70). The majority of RF associated with LF was calculated in the margin around the GTV. Conclusions: Pretherapeutic MRI based RF connected with lesion-specific outcome after SRT could be identified, despite multicentric data and minor differences in imaging protocols. Image data analysis of the surrounding metastatic environment may provide therapy-relevant information with the potential to further individualize radiotherapy strategies.

BackgroundPeristalsis-related streak artifacts on the liver compromise image quality and diagnostic accuracy.PurposeTo assess dual-layer spectral-detector computed tomography (CT) image reconstructions for reducing intestinal peristalsis-related streak artifacts on the liver.MethodsWe retrospectively evaluated 220 contrast-enhanced abdominal dual-energy CT scans in 131 consecutive patients (mean age: 68 ± 10 years, 120 men) who underwent routine clinical dual-layer spectral-detector CT imaging (120 kVp, 40 keV, 200 keV, virtual non-contrast (VNC), iodine images). Two independent readers evaluated bowel peristalsis streak artifacts on the liver qualitatively on a five-point Likert scale (1 = none to 5 = severe) and quantitatively by depth of streak artifact extension into the liver and measurements of Hounsfield Unit and iodine concentration differences from normal liver. Artifact severity between image reconstructions were compared by Wilcoxon signed-rank and paired t-tests.Results12 scans were excluded due to missing spectral data, artifacts on the liver originating from metallic foreign materials, or oral contrast material. Streak artifacts on the liver were seen in 51/208 (25%) scans and involved the left lobe only in 49/51 (96%), the right lobe only in 0/51 (0%), and both lobes in 2/51 (4%) scans. Artifact frequency was lower in iodine than in 120 kVp images (scans 18/208 vs. 51/208, p < 0.001). Artifact severity was less in iodine than in 120 kVp images (median score 1 vs. 3, p < 0.001). Streak artifact extension into the liver was shorter in iodine than 120 kVp images (mean length 2 ± 4 vs. 12 ± 5 mm, p < 0.001). Hounsfield Unit and iodine concentration differed significantly between bright streak artifacts and normal liver in 120 kVp, 40 keV, 200 keV, and VNC images (p < 0.001, each), but not in iodine images (p = 0.23).ConclusionIntestinal peristalsis-related streak artifacts commonly affect the left liver lobe at CT and can be substantially redu

Purpose: To assess the influence of different dual-energy CT (DECT) scanner techniques on the severity of visceral-motion-related artifacts on the liver. Methods: Two independent readers retrospectively evaluated visceral-motion-related artifacts on the liver on 120-kVp(-like), monoenergetic low- and high-keV, virtual non-contrast (VNC), and iodine images acquired on a dual-source, twin-beam, fast kV-switching, and dual-layer spectral detector scanner. Quantitative assessment: Depth of artifact extension into the liver, measurements of Hounsfield Units (HU) and iodine concentrations. Qualitative assessment: Five-point Likert scale (1 = none to 5 = severe). Artifact severity between image reconstructions were compared by Wilcoxon signed-rank and paired t-tests. Results: 615 contrast-enhanced routine clinical DECT scans of the abdomen were evaluated in 458 consecutive patients (mean age: 61 ± 14 years, 331 men). For dual-source and twin-beam scanners, depth of extension of artifacts into the liver was significantly shorter and artifact severity scores significantly lower for 120-kVp-like images compared with the other image reconstructions (p < 0.001, each). For fast kV-switching and spectral detector scanner images, depth of extension of artifacts was significantly shorter and artifact severity scores significantly lower for iodine images (p < 0.001, each). Dual-source 120-kVp-like and spectral detector iodine images reduced artifacts to an extent that no significant difference in HU or iodine concentrations between artifacts (dual-source: 97 HU, spectral detector: 1.9 mg/mL) and unaffected liver parenchyma (dual-source: 108 HU, spectral detector: 2.1 mg/mL) was measurable (dual-source: p = 0.32, spectral detector: p = 0.15). Conclusion: Visceral-motion-related artifacts on the liver can be markedly reduced by viewing 120-kVp-like images for dual-source and twin-beam DECT scanners and iodine images for fast kV-switching and dual-layer spectral detector DECT sca

BackgroundPeristalsis-related streak artifacts on the liver compromise image quality and diagnostic accuracy.PurposeTo assess dual-layer spectral-detector computed tomography (CT) image reconstructions for reducing intestinal peristalsis-related streak artifacts on the liver.MethodsWe retrospectively evaluated 220 contrast-enhanced abdominal dual-energy CT scans in 131 consecutive patients (mean age: 68 ± 10 years, 120 men) who underwent routine clinical dual-layer spectral-detector CT imaging (120 kVp, 40 keV, 200 keV, virtual non-contrast (VNC), iodine images). Two independent readers evaluated bowel peristalsis streak artifacts on the liver qualitatively on a five-point Likert scale (1 = none to 5 = severe) and quantitatively by depth of streak artifact extension into the liver and measurements of Hounsfield Unit and iodine concentration differences from normal liver. Artifact severity between image reconstructions were compared by Wilcoxon signed-rank and paired t-tests.Results12 scans were excluded due to missing spectral data, artifacts on the liver originating from metallic foreign materials, or oral contrast material. Streak artifacts on the liver were seen in 51/208 (25%) scans and involved the left lobe only in 49/51 (96%), the right lobe only in 0/51 (0%), and both lobes in 2/51 (4%) scans. Artifact frequency was lower in iodine than in 120 kVp images (scans 18/208 vs. 51/208, p < 0.001). Artifact severity was less in iodine than in 120 kVp images (median score 1 vs. 3, p < 0.001). Streak artifact extension into the liver was shorter in iodine than 120 kVp images (mean length 2 ± 4 vs. 12 ± 5 mm, p < 0.001). Hounsfield Unit and iodine concentration differed significantly between bright streak artifacts and normal liver in 120 kVp, 40 keV, 200 keV, and VNC images (p < 0.001, each), but not in iodine images (p = 0.23).ConclusionIntestinal peristalsis-related streak artifacts commonly affect the left liver lobe at CT and can be substantially redu

Background: The risk of adverse cardiovascular events in patients with acute myocarditis (AM) and desmosomal gene variants (DGV) remains unknown. Objectives: The purpose of this study was to ascertain the risk of death, ventricular arrhythmias, recurrent myocarditis, and heart failure (main endpoint) in patients with AM and pathogenic or likely pathogenetic DGV. Methods: In a retrospective international study from 23 hospitals, 97 patients were included: 36 with AM and DGV (DGV[+]), 25 with AM and negative gene testing (DGV[−]), and 36 with AM without genetics testing. All patients had troponin elevation plus findings consistent with AM on histology or at cardiac magnetic resonance (CMR). In 86 patients, CMR changes in function and structure were re-assessed at follow-up. Results: In the DGV(+) AM group (88.9% DSP variants), median age was 24 years, 91.7% presented with chest pain, and median left ventricular ejection fraction (LVEF) was 56% on CMR (P = NS vs the other 2 groups). Kaplan-Meier curves demonstrated a higher risk of the main endpoint in DGV(+) AM compared with DGV(−) and without genetics testing patients (62.3% vs 17.5% vs 5.3% at 5 years, respectively; P < 0.0001), driven by myocarditis recurrence and ventricular arrhythmias. At follow-up CMR, a higher number of late gadolinium enhanced segments was found in DGV(+) AM. Conclusions: Patients with AM and evidence of DGV have a higher incidence of adverse cardiovascular events compared with patients with AM without DGV. Further prospective studies are needed to ascertain if genetic testing might improve risk stratification of patients with AM who are considered at low risk.

Despite the success of convolutional neural networks for 3D medical-image segmentation, the architectures currently used are still not robust enough to the protocols of different scanners, and the variety of image properties they produce. Moreover, access to large-scale datasets with annotated regions of interest is scarce, and obtaining good results is thus difficult. To overcome these challenges, we introduce IB-U-Nets, a novel architecture with inductive bias, inspired by the visual processing in vertebrates. With the 3D U-Net as the base, we add two 3D residual components to the second encoder blocks. They provide an inductive bias, helping U-Nets to segment anatomical structures from 3D images with increased robustness and accuracy. We compared IB-U-Nets with state-of-the-art 3D U-Nets on multiple modalities and organs, such as the prostate and spleen, using the same training and testing pipeline, including data processing, augmentation and cross-validation. Our results demonstrate the superior robustness and accuracy of IB-U-Nets, especially on small datasets, as is typically the case in medical-image analysis. IB-U-Nets source code and models are publicly available., Comment: 11 pages, 4 figures

Background: Acute myocarditis (AM) is thought to be a rare cardiovascular complication of COVID-19, although minimal data are available beyond case reports. We aim to report the prevalence, baseline characteristics, in-hospital management, and outcomes for patients with COVID-19-associated AM on the basis of a retrospective cohort from 23 hospitals in the United States and Europe. Methods: A total of 112 patients with suspected AM from 56 963 hospitalized patients with COVID-19 were evaluated between February 1, 2020, and April 30, 2021. Inclusion criteria were hospitalization for COVID-19 and a diagnosis of AM on the basis of endomyocardial biopsy or increased troponin level plus typical signs of AM on cardiac magnetic resonance imaging. We identified 97 patients with possible AM, and among them, 54 patients with definite/probable AM supported by endomyocardial biopsy in 17 (31.5%) patients or magnetic resonance imaging in 50 (92.6%). We analyzed patient characteristics, treatments, and outcomes among all COVID-19-associated AM. Results: AM prevalence among hospitalized patients with COVID-19 was 2.4 per 1000 hospitalizations considering definite/probable and 4.1 per 1000 considering also possible AM. The median age of definite/probable cases was 38 years, and 38.9% were female. On admission, chest pain and dyspnea were the most frequent symptoms (55.5% and 53.7%, respectively). Thirty-one cases (57.4%) occurred in the absence of COVID-19-associated pneumonia. Twenty-one (38.9%) had a fulminant presentation requiring inotropic support or temporary mechanical circulatory support. The composite of in-hospital mortality or temporary mechanical circulatory support occurred in 20.4%. At 120 days, estimated mortality was 6.6%, 15.1% in patients with associated pneumonia versus 0% in patients without pneumonia (P=0.044). During hospitalization, left ventricular ejection fraction, assessed by echocardiography, improved from a median of 40% on admission to 55% at discharge

Pulmonary Tumor Thrombotic Microangiopathy (PTTM) is a rare condition associated with neoplastic disorders, predominantly gastric cancer, leading to pre-capillary Pulmonary Hypertension (PH). The pathologic mechanism involved is a fibrocellular intimal proliferation of small pulmonary vessels sustained by nests of carcinomatous cells lodged in pulmonary vasculature. Clinical presentation is nonspecific, including progressive dyspnea and dry cough. Diagnosis of PTTM is extremely challenging ante-mortem and prognosis is poor. Here we describe the case of a middle-aged man, without known previous cancer history. The clinical course was rapidly unfavorable, with progressive dyspnea and PH associated with hemodynamic instability, eventually culminating in patient's death. PTTM diagnosis was made post-mortem. PTTM should be considered in any patient presenting with unexplained PH, especially if it is rapidly progressive, poorly responsive to standard approaches or there is suspected history of malignancy. A prompt diagnosis of PTTM could help in bringing light into this still under-recognized condition.

Item does not contain fulltext, Current risk-stratification systems for prostate cancer (PCa) do not sufficiently reflect the disease heterogeneity. Genomic classifiers (GC) enable improved risk stratification after surgery, but less data exist for patients treated with definitive radiation therapy (RT) or RT in oligo-/metastatic disease stages. To guide future perspectives of GCs for RT, we conducted (1) a systematic review on the evidence of GCs for patients treated with RT and (2) a survey of experts using the Delphi method, addressing the role of GCs in personalized treatments to identify relevant fields of future clinical and translational research. We performed a systematic review and screened ongoing clinical trials on ClinicalTrials.gov. Based on these results, a multidisciplinary international team of experts received an adapted Delphi method survey. Thirty-one and 30 experts answered round 1 and round 2, respectively. Questions with ≥75% agreement were considered relevant and included in the qualitative synthesis. Evidence for GCs as predictive biomarkers is mainly available to the postoperative RT setting. Validation of GCs as prognostic markers in the definitive RT setting is emerging. Experts used GCs in patients with PCa with extensive metastases (30%), in postoperative settings (27%), and in newly diagnosed PCa (23%). Forty-seven percent of experts do not currently use GCs in clinical practice. Expert consensus demonstrates that GCs are promising tools to improve risk-stratification in primary and oligo-/metastatic patients in addition to existing classifications. Experts were convinced that GCs might guide treatment decisions in terms of RT-field definition and intensification/deintensification in various disease stages. This work confirms the value of GCs and the promising evidence of GC utility in the setting of RT. Additional studies of GCs as prognostic biomarkers are anticipated and form the basis for future studies addressing predictive capabilities of GCs to optimize RT and

The challenge of global food security in the face of population growth, conflict and climate change requires a comprehensive understanding of cropped areas, irrigation practices and the distribution of major commodity crops like maize and wheat. However, such understanding should preferably be updated at seasonal intervals for each agricultural system rather than relying on a single annual assessment. Here we present the European Space Agency funded WorldCereal system, a global, seasonal, and reproducible crop and irrigation mapping system that addresses existing limitations in current global-scale crop and irrigation mapping. WorldCereal generates a range of global products, including temporary crop extent, seasonal maize and cereals maps, seasonal irrigation maps, seasonal active cropland maps, and confidence layers providing insights into expected product quality. The WorldCereal product suite for the year 2021 presented here serves as a global demonstration of the dynamic open-source WorldCereal system. The presented products are fully validated, e.g., global user's and producer's accuracies for the annual temporary crop product are 88.5 % and 92.1 %, respectively. The WorldCereal system provides a vital tool for policymakers, international organizations, and researchers to better understand global crop and irrigation patterns and inform decision-making related to food security and sustainable agriculture. Our findings highlight the need for continued community efforts such as additional reference data collection to support further development and push the boundaries for global agricultural mapping from space. The global products are available at https://doi.org/10.5281/zenodo.7875104 (Van Tricht et al., 2023).

The challenge of global food security in the face of population growth, conflict and climate change requires a comprehensive understanding of cropped areas, irrigation practices and the distribution of major commodity crops like maize and wheat. However, such understanding should preferably be updated at seasonal intervals for each agricultural system rather than relying on a single annual assessment. Here we present the European Space Agency funded WorldCereal system, a global, seasonal, and reproducible crop and irrigation mapping system that addresses existing limitations in current global-scale crop and irrigation mapping. WorldCereal generates a range of global products, including temporary crop extent, seasonal maize and cereals maps, seasonal irrigation maps, seasonal active cropland maps, and confidence layers providing insights into expected product quality. The WorldCereal product suite for the year 2021 presented here serves as a global demonstration of the dynamic open-source WorldCereal system. The presented products are fully validated, e.g., global user's and producer's accuracies for the annual temporary crop product are 88.5 % and 92.1 %, respectively. The WorldCereal system provides a vital tool for policymakers, international organizations, and researchers to better understand global crop and irrigation patterns and inform decision-making related to food security and sustainable agriculture. Our findings highlight the need for continued community efforts such as additional reference data collection to support further development and push the boundaries for global agricultural mapping from space. The global products are available at https://doi.org/10.5281/zenodo.7875104 (Van Tricht et al., 2023).

The assumption of perfect knowledge of rate parameters in continuous-time Markov chains (CTMCs) is undermined when confronted with reality, where they may be uncertain due to lack of information or because of measurement noise. Here we consider uncertain CTMCs (UCTMCs), where rates are assumed to vary non-deterministically with time from bounded continuous intervals. An uncertain CTMC can be therefore seen as a specific type of Markov decision process for which the analysis is computationally difficult. To tackle this, we develop a theory of minimization which generalizes the notion of lumpability for CTMCs. Our first result is a quantitative and logical characterization of minimization. Specifically, we show that the reduced UCTMC model has a macro-state for each block of a partition of the state space, which preserves value functions and logical formulae whenever rewards are equal within each block. The second result is an efficient minimization algorithm for UCTMCs by means of partition refinement. As application, we show that reductions in a number of CTMCbenchmark models are robust with respect to uncertainties in original rates.

Purpose: Tumor hypoxia is a paradigmatic negative prognosticator of treatment resistance in head and neck squamous cell carcinoma (HNSCC). The lack of robust and reliable hypoxia classifiers limits the adaptation of stratified therapies. We hypothesized that the tumor DNA methylation landscape might indicate epigenetic reprogramming induced by chronic intratumoral hypoxia. Experimental Design: A DNA-methylome–based tumor hypoxia classifier (Hypoxia-M) was trained in the TCGA (The Cancer Genome Atlas)-HNSCC cohort based on matched assignments using gene expression–based signatures of hypoxia (Hypoxia-GES). Hypoxia-M was validated in a multicenter DKTK-ROG trial consisting of human papillomavirus (HPV)–negative patients with HNSCC treated with primary radiochemotherapy (RCHT). Results: Although hypoxia-GES failed to stratify patients in the DKTK-ROG, Hypoxia-M was independently prognostic for local recurrence (HR, 4.3; P ¼0.001) and overall survival (HR, 2.34; P ¼ 0.03) but not distant metastasis after RCHT in both cohorts. Hypoxia-M status was inversely associated with CD8 T-cell infiltration in both cohorts. Hypoxia-M was further prognostic in the TCGA-PanCancer cohort (HR, 1.83; P ¼0.04), underscoring the breadth of this classifier for predicting tumor hypoxia status. Conclusions: Our findings highlight an unexplored avenue for DNA methylation–based classifiers as biomarkers of tumoral hypoxia for identifying high-risk features in patients with HNSCC tumors.

Thermal management protects against external agents and increases the lifetime and performance of the devices in which it is implemented. Because of their ability to store and release a high amount of energy at a nearly constant temperature, phase change materials (PCMs) are promising thermoregulatory materials. Thus, the manufacture of PVDF fibers containing PCMs has advantages since PVDF is already used in elements that are susceptible to thermal management as a binder in batteries or as a base material for fabrics. This work presents a simple, versatile, in situ, cost-effective, and easy-to-scale-up method to produce PVDF-based fibers containing paraffin RT-28HC for thermal management. To achieve that goal, the microfluidic approach of coaxial flows was simplified to gravity-aided laminar jet injection into a bulk fluid, where fibers were produced by the solvent extraction mechanism. With this methodology, hollow PVDF fibers and core-shell PVDF fibers containing paraffin RT-28HC have been produced. The proposed approach resulted in fibers with up to 98 J/g of latent heat, with a hierarchical porous structure. SEM study of the fiber morphology has shown that PCM is in the form of slugs along the fibers. Such PCM distribution is maintained until the first melting cycle, when molten PCM spreads within the fiber under capillary forces, which was observed by an infrared camera. Manufactured composite fibers have shown low thermal conductivity and high elasticity, which suggest their potential application as a thermal insulation material with thermal buffer properties. Leakage tests revealed outstanding retention capacity with only 3.5% mass loss after 1000 melting/crystallization cycles. Finally, tensile tests were carried out to evaluate the mechanical properties of the fibers before and after thermal cycling.

The challenge of global food security in the face of population growth, conflict and climate change requires a comprehensive understanding of cropped areas, irrigation practices and the distribution of major commodity crops like maize and wheat. However, such understanding should preferably be updated at seasonal intervals for each agricultural system rather than relying on a single annual assessment. Here we present the European Space Agency funded WorldCereal system, a global, seasonal, and reproducible crop and irrigation mapping system that addresses existing limitations in current global-scale crop and irrigation mapping. WorldCereal generates a range of global products, including temporary crop extent, seasonal maize and cereals maps, seasonal irrigation maps, seasonal active cropland maps, and confidence layers providing insights into expected product quality. The WorldCereal product suite for the year 2021 presented here serves as a global demonstration of the dynamic open-source WorldCereal system. The presented products are fully validated, e.g., global user's and producer's accuracies for the annual temporary crop product are 88.5 % and 92.1 %, respectively. The WorldCereal system provides a vital tool for policymakers, international organizations, and researchers to better understand global crop and irrigation patterns and inform decision-making related to food security and sustainable agriculture. Our findings highlight the need for continued community efforts such as additional reference data collection to support further development and push the boundaries for global agricultural mapping from space. The global products are available at https://doi.org/10.5281/zenodo.7875104 (Van Tricht et al., 2023).

Background Accurate estimation of the effective reproductive number ( $$R_e$$ R e ) of epidemic outbreaks is of central relevance to public health policy and decision making. We present estimateR, an R package for the estimation of the reproductive number through time from delayed observations of infection events. Such delayed observations include confirmed cases, hospitalizations or deaths. The package implements the methodology of Huisman et al. but modularizes the $$R_e$$ R e estimation procedure to allow easy implementation of new alternatives to the currently available methods. Users can tailor their analyses according to their particular use case by choosing among implemented options. Results The estimateR R package allows users to estimate the effective reproductive number of an epidemic outbreak based on observed cases, hospitalization, death or any other type of event documenting past infections, in a fast and timely fashion. We validated the implementation with a simulation study: estimateR yielded estimates comparable to alternative publicly available methods while being around two orders of magnitude faster. We then applied estimateR to empirical case-confirmation incidence data for COVID-19 in nine countries and for dengue fever in Brazil; in parallel, estimateR is already being applied (i) to SARS-CoV-2 measurements in wastewater data and (ii) to study influenza transmission based on wastewater and clinical data in other studies. In summary, this R package provides a fast and flexible implementation to estimate the effective reproductive number for various diseases and datasets. Conclusions The estimateR R package is a modular and extendable tool designed for outbreak surveillance and retrospective outbreak investigation. It extends the method developed for COVID-19 by Huisman et al. and makes it available for a variety of pathogens, outbreak scenarios, and observation types. Estimates obtained with estimateR can be interpreted directly or use

Tension-type headache (TTH) and migraine are two common primary headaches distinguished by clinical characteristics according to the 3rd edition of the International Classification of Headache Disorders. Migraine is identified by specific features such as being more prevalent in females, being aggravated by physical activity, certain genetic factors, having photophobia, phonophobia, nausea, vomiting, or aura, and responding to specific drugs. Nonetheless, TTH and migraine share some common characteristics, such as onset occurring in the 20 s, and being triggered by psychological factors like stress, moderate pain severity, and mild nausea in chronic TTH. Both conditions involve the trigeminovascular system in their pathophysiology. However, distinguishing between TTH and migraine in clinical practice, research, and epidemiological studies can be challenging, as there is a lack of specific diagnostic tests and biomarkers. Moreover, both conditions may coexist, further complicating the diagnostic process. This review aims to explore the similarities and differences in the pathophysiology, epidemiology, burden and disability, comorbidities, and responses to pharmacological and non-pharmacological treatments of TTH and migraine. The review also discusses future research directions to address the diagnostic challenges and improve the understanding and management of these conditions. Graphical Abstract: [Figure not available: see fulltext.]

Aim: According to the Latitudinal Biotic Interaction Hypothesis (LBIH), the general increase in biodiversity towards lower latitudes can be partially explained by an increase in the intensity of biotic interactions. While LBIH received some support for plant-herbivores interactions, much less is known about how higher trophic levels may contribute to shape biotic interactions across latitudinal gradients. We hypothesized that the diversity of insectivorous birds increases towards lower latitude, leading to higher predation rates on insect herbivores. Location: Europe. Taxon: Insectivorous birds and pedunculate oaks. Methods: We deployed plasticine caterpillars in 138 oak trees in 47 sites along a 19° latitudinal gradient in Europe to quantify bird insectivory through predation attempts. In addition, we used passive acoustic monitoring to (i) characterize the acoustic diversity of surrounding soundscapes; and (ii) infer both taxonomic and functional diversity of insectivorous birds from recordings. Results: The functional diversity of insectivorous birds increased towards lower latitude. Bird predation increased with latitude, forest cover and bird acoustic diversity but decreased with mean annual temperature and functional richness of insectivorous birds. Contrary to our predictions, latitudinal clines in bird predation attempts were not directly mediated by changes in insectivorous bird diversity or acoustic diversity, but latitude and habitat still had independent effects on predation attempts. Main conclusions: Our study does not fully support the predictions of the LBIH of more biotic interactions southwards and advocates for better accounting for activity and abundance of insectivorous birds when studying the large-scale variation in insect-tree interactions.

The European Space Agency (ESA) WorldCereal 10m 2021 product suite consist of global-scale annual and seasonal crop maps and (where applicable) their related confidence. Every file in this repository contains up to 106 agro-ecological zone (AEZ) products which were all processed with respect to their own regional seasonality and should be considered as independent products.

The challenge of global food security in the face of population growth, conflict and climate change requires a comprehensive understanding of cropped areas, irrigation practices and the distribution of major commodity crops like maize and wheat. However, such understanding should preferably be updated at seasonal intervals for each agricultural system rather than relying on a single annual assessment. Here we present the European Space Agency funded WorldCereal system, a global, seasonal, and reproducible crop and irrigation mapping system that addresses existing limitations in current global-scale crop and irrigation mapping. WorldCereal generates a range of global products, including temporary crop extent, seasonal maize and cereals maps, seasonal irrigation maps, seasonal active cropland maps, and confidence layers providing insights into expected product quality. The WorldCereal product suite for the year 2021 presented here serves as a global demonstration of the dynamic open-source WorldCereal system. The presented products are fully validated, e.g., global user's and producer's accuracies for the annual temporary crop product are 88.5 % and 92.1 %, respectively. The WorldCereal system provides a vital tool for policymakers, international organizations, and researchers to better understand global crop and irrigation patterns and inform decision-making related to food security and sustainable agriculture. Our findings highlight the need for continued community efforts such as additional reference data collection to support further development and push the boundaries for global agricultural mapping from space. The global products are available at https://doi.org/10.5281/zenodo.7875104 (Van Tricht et al., 2023).

The approximation capability of ANNs and their RNN instantiations, is strongly correlated with the number of parameters packed into these networks. However, the complexity barrier for human understanding, is arguably related to the number of neurons and synapses in the networks, and to the associated nonlinear transformations. In this paper we show that the use of biophysical synapses, as found in LTCs, have two main benefits. First, they allow to pack more parameters for a given number of neurons and synapses. Second, they allow to formulate the nonlinear-network transformation, as a linear system with state-dependent coefficients. Both increase interpretability, as for a given task, they allow to learn a system linear in its input features, that is smaller in size compared to the state of the art. We substantiate the above claims on various time-series prediction tasks, but we believe that our results are applicable to any feedforward or recurrent ANN.

Exoskeletons are becoming the reference technology for assistance and augmentation of human motor functions in a wide range of application domains. Unfortunately, the exponential growth of this sector has not been accompanied by a rigorous risk assessment (RA) process, which is necessary to identify the major aspects concerning the safety and impact of this new technology on humans. This situation may seriously hamper the market uptake of new products. This paper presents the results of a survey that was circulated to understand how hazards are considered by exoskeleton users, from research and industry perspectives. Our analysis aimed to identify the perceived occurrence and the impact of a sample of generic hazards, as well as to collect suggestions and general opinions from the respondents that can serve as a reference for more targeted RA. Our results identified a list of relevant hazards for exoskeletons. Among them, misalignments and unintended device motion were perceived as key aspects for exoskeletons’ safety. This survey aims to represent a first attempt in recording overall feedback from the community and contribute to future RAs and the identification of better mitigation strategies in the field.

Evaluating human-exoskeleton interaction typically requires experiments with human subjects, which raises safety issues and entails time-consuming testing procedures. This paper presents a mechatronic replica of a human leg, which was designed to quantify physical interaction dynamics between exoskeletons and human limbs without the need for human testing. In the first part of this work, we present the mechanical, electronic, sensory system and software solutions integrated in our leg replica prototype. In the second part, we used the leg replica to test its interaction with two types of commercially available wearable devices, i.e. an active full leg exoskeleton and a passive knee orthosis. We ran basic test examples to demonstrate the functioning and benchmarking potential of the leg replica to assess the effects of joint misalignments on force transmission. The integrated force sensors embedded in the leg replica detected higher interaction forces in the misaligned scenario in comparison to the aligned one, in both active and passive modalities. The small standard deviation of force measurements across cycles demonstrates the potential of the leg replica as a standard test method for reproducible studies of human-exoskeleton physical interaction.

The present invention is related to an anatomical model for measuring the interaction of an exoskeleton with a biological limb assisted by said exoskeleton. The present invention is characterized by a special structure of the device formed by an inner structure comprising a first structural element and a second structural element. The first structural element and a second structural element are joined by means of a first joint, preferably with rotation capability. Likewise, the device comprises a first casing and a second casing, each of the casings covering the first structural element and a second structural element respectively. The device is further characterized in that each of the shells is divided into a plurality of segments and, each segment is attached to the internal structure by means of a sensor that allows to provide a signal representing the loads that the exoskeleton exerts on the region corresponding to that segment.

[EN] The present invention is related to an anatomical model for measuring the interaction of an exoskeleton with a biological limb assisted by said exoskeleton. The present invention is characterized by a special structure of the device formed by an inner structure comprising a first structural element and a second structural element. The first structural element and a second structural element are joined by means of a first joint, preferably with rotation capability. Likewise, the device comprises a first casing and a second casing, each of the casings covering the first structural element and a second structural element respectively. The device is further characterized in that each of the shells is divided into a plurality of segments and, each segment is attached to the internal structure by means of a sensor that allows to provide a signal representing the loads that the exoskeleton exerts on the region corresponding to that segment., [FR] La présente invention concerne un modèle anatomique servant à mesurer l'interaction entre un exosquelette et un membre biologique assisté par ledit exosquelette. La présente invention est caractérisée par une structure spéciale du dispositif formée par une structure interne comprenant un premier élément structural et un second élément structural. Le premier élément structural et un second élément structural sont assemblés au moyen d'un premier joint, de préférence de manière à pouvoir tourner. De même, le dispositif comprend une première enveloppe et une seconde enveloppe, chacune des enveloppes recouvrant respectivement le premier élément structural et un second élément structural. Le dispositif est en outre caractérisé en ce que chacune des enveloppes est divisée en une pluralité de segments, et chaque segment est fixé à la structure interne au moyen d'un capteur qui permet de fournir un signal représentant les charges que l'exosquelette exerce sur la région correspondant à ce segment.

The challenge of global food security in the face of population growth, conflict, and climate change requires a comprehensive understanding of cropped areas, irrigation practices, and the distribution of major commodity crops like maize and wheat. However, such understanding should preferably be updated at seasonal intervals for each agricultural system rather than relying on a single annual assessment. Here we present the European Space Agency-funded WorldCereal system, a global, seasonal, and reproducible crop and irrigation mapping system that addresses existing limitations in current global-scale crop and irrigation mapping. WorldCereal generates a range of global products, including temporary crop extent, seasonal maize and cereal maps, seasonal irrigation maps, seasonal active cropland maps, and model confidence layers providing insights into expected product quality. The WorldCereal product suite for the year 2021 presented here serves as a global demonstration of the dynamic open-source WorldCereal system. Validation of the products was done based on best available reference data per product. A global statistical validation for the temporary crop extent product resulted in user’s and producer’s accuracies of 88.5 % and 92.1 %, respectively. For crop type, a verification was performed against a newly collected street view dataset (overall agreement 82.5 %) and a limited number of publicly available in situ datasets (reaching minimum agreement of 80 %). Finally, global irrigated-area estimates were derived from available maps and statistical datasets, revealing the conservative nature of the WorldCereal irrigation product. The WorldCereal system provides a vital tool for policymakers, international organizations, and researchers to better understand global crop and irrigation patterns and to inform decision-making related to food security and sustainable agriculture. Our findings highlight the need for continued community efforts such as additional reference d

Genetic alterations help predict the clinical behavior of diffuse gliomas, but some variability remains uncorrelated. Here, we demonstrate that haploinsufficient deletions of chromatin-bound tumor suppressor NFKB inhibitor alpha (NFKBIA) display distinct patterns of occurrence in relation to other genetic markers and are disproportionately present at recurrence. NFKBIA haploinsufficiency is associated with unfavorable patient outcomes, independent of genetic and clinicopathologic predictors. NFKBIA deletions reshape the DNA and histone methylome antipodal to the IDH mutation and induce a transcriptome landscape partly reminiscent of H3K27M mutant pediatric gliomas. In IDH mutant gliomas, NFKBIA deletions are common in tumors with a clinical course similar to that of IDH wild-type tumors. An externally validated nomogram model for estimating individual patient survival in IDH mutant gliomas confirms that NFKBIA deletions predict comparatively brief survival. Thus, NFKBIA haploinsufficiency aligns with distinct epigenome changes, portends a poor prognosis, and should be incorporated into models predicting the disease fate of diffuse gliomas.

Biologically younger, fully independent octogenarians are able to tolerate most oncological treatments. Increasing frailty results in decreasing eligibility for certain treatments, e.g., chemotherapy and surgery. Most brain metastases are not an isolated problem, but part of widespread cancer dissemination, often in combination with compromised performance status. Multidisciplinary assessment is key in this vulnerable patient population where age, frailty, comorbidity and even moderate additional deficits from brain metastases or their treatment may result in immobilization, hospitalization, need for nursing home care, termination of systemic anticancer treatment etc. Here, we provide examples of successful treatment (surgery, radiosurgery, systemic therapy) and best supportive care, and comment on the limitations of prognostic scores, which often were developed in all-comers rather than octogenarians. Despite selection bias in retrospective studies, survival after radiosurgery was more encouraging than after whole-brain radiotherapy. Prospective research with focus on octogenarians is warranted to optimize outcomes.

BACKGROUND Tumor hypoxia is a paradigmatic negative prognosticator of treatment resistance in Head and Neck Squamous Cell Carcinoma (HNSCC). The lack of robust and reliable hypoxia classifiers limits the adaptation of stratified therapies. We hypothesized that the tumor DNA methylation landscape might indicate epigenetic reprogramming induced by chronic intratumoral hypoxia. METHODS A DNA methylome-based tumor hypoxia classifier (Hypoxia-M) was trained in the TCGA-HNSCC cohort based on matched assignments using gene expression-based signatures of hypoxia (Hypoxia-GES). Hypoxia-M was validated in a multicenter DKTK-ROG trial consisting of Human Papilloma Virus (HPV)-negative HNSCC patients treated with primary radiochemotherapy (RCHT). RESULTS While hypoxia-GSEs failed to stratify patients in the DKTK-ROG, Hypoxia-M was independently prognostic for local recurrence (LR, HR=4.3, p=0.001) and overall survival (OS, HR=2.34, p=0.03) but not distant metastasis (DM) after RCHT in the both cohorts. Hypoxia-M status was inversely associated with CD8 T-cells infiltration in both cohorts. Hypoxia-M was further prognostic in the TCGA-PanCancer cohort (HR=1.83, p=0.04), underscoring the breadth of this classifier for predicting tumor hypoxia status. CONCLUSIONS Our findings highlight an unexplored avenue for DNA Methylation-based classifiers as biomarkers of tumoral hypoxia for identifying high-risk features in patients with HNSCC tumors. TRIAL REGISTRATION Retrospective observational study from the German Cancer Consortium (DKTK-ROG), not interventional.

BACKGROUND Many automatic approaches to brain tumor segmentation employ multiple magnetic resonance imaging (MRI) sequences. The goal of this project was to compare different combinations of input sequences to determine which MRI sequences are needed for effective automated brain metastasis (BM) segmentation. METHODS We analyzed preoperative imaging (T1-weighted sequence ± contrast-enhancement (T1/T1-CE), T2-weighted sequence (T2), and T2 fluid-attenuated inversion recovery (T2-FLAIR) sequence) from 339 patients with BMs from seven centers. A baseline 3D U-Net with all four sequences and six U-Nets with plausible sequence combinations (T1-CE, T1, T2-FLAIR, T1-CE + T2-FLAIR, T1-CE + T1 + T2-FLAIR, T1-CE + T1) were trained on 239 patients from two centers and subsequently tested on an external cohort of 100 patients from five centers. RESULTS The model based on T1-CE alone achieved the best segmentation performance for BM segmentation with a median Dice similarity coefficient (DSC) of 0.96. Models trained without T1-CE performed worse (T1-only: DSC = 0.70 and T2-FLAIR-only: DSC = 0.73). For edema segmentation, models that included both T1-CE and T2-FLAIR performed best (DSC = 0.93), while the remaining four models without simultaneous inclusion of these both sequences reached a median DSC of 0.81-0.89. CONCLUSIONS A T1-CE-only protocol suffices for the segmentation of BMs. The combination of T1-CE and T2-FLAIR is important for edema segmentation. Missing either T1-CE or T2-FLAIR decreases performance. These findings may improve imaging routines by omitting unnecessary sequences, thus allowing for faster procedures in daily clinical practice while enabling optimal neural network-based target definitions.

BACKGROUND AND OBJECTIVES The validated LabBM score (laboratory parameters in patients with brain metastases) represents a widely applicable survival prediction model, which incorporates 5 blood test results (serum lactate dehydrogenase (LDH), C-reactive protein (CRP), albumin, platelets and hemoglobin). All tests are classified as normal or abnormal, without accounting for the wide range of abnormality observed in practice. We tested the hypothesis that improved stratification might be possible, if more granular test results are employed. METHODS Retrospective analysis of 198 patients managed with primary whole-brain radiotherapy in one of the institutions who validated the original LabBM score. RESULTS For two blood tests (albumin, CRP), discrimination was best for the original dichotomized version (normal/abnormal). For two others (LDH, hemoglobin), a three-tiered classification was best. The number of patients with low platelet count was not large enough for detailed analyses. A modified LabBM score was developed, which separates the intermediate of originally 3 prognostic groups into 2 statistically significantly different strata, resulting in a 4-tiered score. CONCLUSION This initial proof-of-principle study suggests that granular blood test results might contribute to further improvement of the score, or alternatively development of a nomogram, if additional large-scale studies confirm the encouraging results of the present analysis.

IMPORTANCE Prostate-specific antigen membrane positron-emission tomography (PSMA-PET) is increasingly used to guide salvage radiotherapy (sRT) after radical prostatectomy for patients with recurrent or persistent prostate cancer. OBJECTIVE To develop and validate a nomogram for prediction of freedom from biochemical failure (FFBF) after PSMA-PET-based sRT. DESIGN, SETTING, AND PARTICIPANTS This retrospective cohort study included 1029 patients with prostate cancer treated between July 1, 2013, and June 30, 2020, at 11 centers from 5 countries. The initial database consisted of 1221 patients. All patients had a PSMA-PET scan prior to sRT. Data were analyzed in November 2022. EXPOSURES Patients with a detectable post-radical prostatectomy prostate-specific antigen (PSA) level treated with sRT to the prostatic fossa with or without additional sRT to pelvic lymphatics or concurrent androgen deprivation therapy (ADT) were eligible. MAIN OUTCOMES AND MEASURES The FFBF rate was estimated, and a predictive nomogram was generated and validated. Biochemical relapse was defined as a PSA nadir of 0.2 ng/mL after sRT. RESULTS In the nomogram creation and validation process, 1029 patients (median age at sRT, 70 years [IQR, 64-74 years]) were included and further divided into a training set (n = 708), internal validation set (n = 271), and external outlier validation set (n = 50). The median follow-up was 32 months (IQR, 21-45 months). Based on the PSMA-PET scan prior to sRT, 437 patients (42.5%) had local recurrences and 313 patients (30.4%) had nodal recurrences. Pelvic lymphatics were electively irradiated for 395 patients (38.4%). All patients received sRT to the prostatic fossa: 103 (10.0%) received a dose of less than 66 Gy, 551 (53.5%) received a dose of 66 to 70 Gy, and 375 (36.5%) received a dose of more than 70 Gy. Androgen deprivation therapy was given to 325 (31.6%) patients. On multivariable Cox proportional hazards regression analysis, pre-sRT PSA level (hazard ratio

Tension-type headache (TTH) and migraine are two common primary headaches distinguished by clinical characteristics according to the 3rd edition of the International Classification of Headache Disorders. Migraine is identified by specific features such as being more prevalent in females, being aggravated by physical activity, certain genetic factors, having photophobia, phonophobia, nausea, vomiting, or aura, and responding to specific drugs. Nonetheless, TTH and migraine share some common characteristics, such as onset occurring in the 20 s, and being triggered by psychological factors like stress, moderate pain severity, and mild nausea in chronic TTH. Both conditions involve the trigeminovascular system in their pathophysiology. However, distinguishing between TTH and migraine in clinical practice, research, and epidemiological studies can be challenging, as there is a lack of specific diagnostic tests and biomarkers. Moreover, both conditions may coexist, further complicating the diagnostic process. This review aims to explore the similarities and differences in the pathophysiology, epidemiology, burden and disability, comorbidities, and responses to pharmacological and non-pharmacological treatments of TTH and migraine. The review also discusses future research directions to address the diagnostic challenges and improve the understanding and management of these conditions. Graphical Abstract: [Figure not available: see fulltext.]

Current risk-stratification systems for prostate cancer (PCa) do not sufficiently reflect the disease heterogeneity. Genomic classifiers (GC) enable improved risk stratification after surgery, but less data exist for patients treated with definitive radiation therapy (RT) or RT in oligo-/metastatic disease stages. To guide future perspectives of GCs for RT, we conducted (1) a systematic review on the evidence of GCs for patients treated with RT and (2) a survey of experts using the Delphi method, addressing the role of GCs in personalized treatments to identify relevant fields of future clinical and translational research. We performed a systematic review and screened ongoing clinical trials on ClinicalTrials.gov. Based on these results, a multidisciplinary international team of experts received an adapted Delphi method survey. Thirty-one and 30 experts answered round 1 and round 2, respectively. Questions with ?75% agreement were considered relevant and included in the qualitative synthesis. Evidence for GCs as predictive biomarkers is mainly available to the postoperative RT setting. Validation of GCs as prognostic markers in the definitive RT setting is emerging. Experts used GCs in patients with PCa with extensive metastases (30%), in postoperative settings (27%), and in newly diagnosed PCa (23%). Forty-seven percent of experts do not currently use GCs in clinical practice. Expert consensus demonstrates that GCs are promising tools to improve risk-stratification in primary and oligo-/metastatic patients in addition to existing classifications. Experts were convinced that GCs might guide treatment decisions in terms of RT-field definition and intensification/deintensification in various disease stages. This work confirms the value of GCs and the promising evidence of GC utility in the setting of RT. Additional studies of GCs as prognostic biomarkers are anticipated and form the basis for future studies addressing predictive capabilities of GCs to optimize RT and, European Union (EU); Horizon 2020; Research and Innovation Programme; German Federal Ministry of Education and Research (BMBF); ERA PER Joint Funding Call 2019; A.T. acknowledges support from Cancer Research UK (C33589/A28284 and C7224/A28724 CRUK RadNet). This project represents independent research supported by the National Institute for Health Research (NIHR) Biomedical Research Centre at the Royal Marsden NHS Foundation Trust and the Institute of Cancer Research, London. The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care. We thank Fotini Miltiadous for administrative support. ; Funding text 3: Disclosures: A.U.K. reports funding support from grant P50CA09213 from the Prostate Cancer National Institutes of Health Specialized Programs of Research Excellence and grant W81XWH-22-1-0044 from the Department of Defense, as well as grant RSD1836 from the Radiologic Society of North America, the STOP Cancer organization, the Jonsson Comprehensive Cancer Center, and the Prostate Cancer Foundation. C.Z. received funding from the Klaus Tschira foundation, Naslund Medical, and from the German Research Foundation. C.Z. received speaker fees from Johnson & Johnson and Novocure, all outside the submitted work. E.D. is an employee of Veracyte, manufacturer of Decipher. A.R. is a consultant and/or speaker for Astellas, Bayer, Pfizer, Blue Earth, Lantheus, Janssen, Tempus, and Veracyte. X.G. is on the consulting/advisory board for Bayer, Myovant, and Guardant Health.

Bio-inspired neural networks have the potential to advance our understanding of neural computation and improve the state-of-the-art of AI systems. Bio-electrical synapses directly transmit neural signals, by enabling fast current flow between neurons. In contrast, bio-chemical synapses transmit neural signals indirectly, through neurotransmitters. Prior work showed that interpretable dynamics for complex robotic control, can be achieved by using chemical synapses, within a sparse, bio-inspired architecture, called Neural Circuit Policies (NCPs). However, a comparison of these two synaptic models, within the same architecture, remains an unexplored area. In this work we aim to determine the impact of using chemical synapses compared to electrical synapses, in both sparse and all-to-all connected networks. We conduct experiments with autonomous lane-keeping through a photorealistic autonomous driving simulator to evaluate their performance under diverse conditions and in the presence of noise. The experiments highlight the substantial influence of the architectural and synaptic-model choices, respectively. Our results show that employing chemical synapses yields noticeable improvements compared to electrical synapses, and that NCPs lead to better results in both synaptic models.

In this paper we propose real-time recurrent reinforcement learning (RTRRL), a biologically plausible approach to solving discrete and continuous control tasks in partially-observable markov decision processes (POMDPs). RTRRL consists of three parts: (1) a Meta-RL RNN architecture, implementing on its own an actor-critic algorithm; (2) an outer reinforcement learning algorithm, exploiting temporal difference learning and dutch eligibility traces to train the Meta-RL network; and (3) random-feedback local-online (RFLO) learning, an online automatic differentiation algorithm for computing the gradients with respect to parameters of the network.Our experimental results show that by replacing the optimization algorithm in RTRRL with the biologically implausible back propagation through time (BPTT), or real-time recurrent learning (RTRL), one does not improve returns, while matching the computational complexity for BPTT, and even increasing complexity for RTRL. RTRRL thus serves as a model of learning in biological neural networks, mimicking reward pathways in the basal ganglia., Comment: 14 pages, 9 figures, includes Appendix

Ensuring safety in dynamic multi-agent systems is challenging due to limited information about the other agents. Control Barrier Functions (CBFs) are showing promise for safety assurance but current methods make strong assumptions about other agents and often rely on manual tuning to balance safety, feasibility, and performance. In this work, we delve into the problem of adaptive safe learning for multi-agent systems with CBF. We show how emergent behavior can be profoundly influenced by the CBF configuration, highlighting the necessity for a responsive and dynamic approach to CBF design. We present ASRL, a novel adaptive safe RL framework, to fully automate the optimization of policy and CBF coefficients, to enhance safety and long-term performance through reinforcement learning. By directly interacting with the other agents, ASRL learns to cope with diverse agent behaviours and maintains the cost violations below a desired limit. We evaluate ASRL in a multi-robot system and a competitive multi-agent racing scenario, against learning-based and control-theoretic approaches. We empirically demonstrate the efficacy and flexibility of ASRL, and assess generalization and scalability to out-of-distribution scenarios. Code and supplementary material are public online., Comment: Update with appendix

Modern tourism in the 21st century is facing numerous challenges. Among these the rapidly growing number of tourists visiting space-limited regions like historical cities, museums and bottlenecks such as bridges is one of the biggest. In this context, a proper and accurate prediction of tourism volume and tourism flow within a certain area is important and critical for visitor management tasks such as sustainable treatment of the environment and prevention of overcrowding. Static flow control methods like conventional low-level controllers or limiting access to overcrowded venues could not solve the problem yet. In this paper, we empirically evaluate the performance of state-of-the-art deep-learning methods such as RNNs, GNNs, and Transformers as well as the classic statistical ARIMA method. Granular limited data supplied by a tourism region is extended by exogenous data such as geolocation trajectories of individual tourists, weather and holidays. In the field of visitor flow prediction with sparse data, we are thereby capable of increasing the accuracy of our predictions, incorporating modern input feature handling as well as mapping geolocation data on top of discrete POI data., Comment: Accepted for publication at the proceedings of the 5th International Data Science Conference - iDSC2023. arXiv admin note: substantial text overlap with arXiv:2206.13274

Robust and efficient learning remains a challenging problem in robotics, in particular with complex visual inputs. Inspired by human attention mechanism, with which we quickly process complex visual scenes and react to changes in the environment, we think that embedding auxiliary information about focus point into robot learning would enhance efficiency and robustness of the learning process. In this paper, we propose a novel approach to model and emulate the human attention with an approximate prediction model. We then leverage this output and feed it as a structured auxiliary feature map into downstream learning tasks. We validate this idea by learning a prediction model from human-gaze recordings of manual driving in the real world. We test our approach on two learning tasks - object detection and imitation learning. Our experiments demonstrate that the inclusion of predicted human attention leads to improved robustness of the trained models to out-of-distribution samples and faster learning in low-data regime settings. Our work highlights the potential of incorporating structured auxiliary information in representation learning for robotics and opens up new avenues for research in this direction. All code and data are available online., Comment: This work has been accepted for the RAP4Robots workshop at ICRA 2023 in London

This paper presents, in a unified fashion, deterministic as well as statistical Lagrangian-verification techniques. They formally quantify the behavioral robustness of any time-continuous process, formulated as a continuous-depth model. To this end, we review LRT-NG, SLR, and GoTube, algorithms for constructing a tight reachtube, that is, an over-approximation of the set of states reachable within a given time-horizon, and provide guarantees for the reachtube bounds. We compare the usage of the variational equations, associated to the system equations, the mean value theorem, and the Lipschitz constants, in achieving deterministic and statistical guarantees. In LRT-NG, the Lipschitz constant is used as a bloating factor of the initial perturbation, to compute the radius of an ellipsoid in an optimal metric, which over-approximates the set of reachable states. In SLR and GoTube, we get statistical guarantees, by using the Lipschitz constants to compute local balls around samples. These are needed to calculate the probability of having found an upper bound, of the true maximum perturbation at every timestep. Our experiments demonstrate the superior performance of Lagrangian techniques, when compared to LRT, Flow*, and CAPD, and illustrate their use in the robustness analysis of various continuous-depth models., Comment: arXiv admin note: text overlap with arXiv:2107.08467