Search

In recent years, advancements in the field of artificial intelligence (AI) have impacted several areas of research and application. Besides more prominent examples like self-driving cars or media consumption algorithms, AI-based systems have further started to gain more and more popularity in the health care sector, however whilst being restrained by high requirements for accuracy, robustness, and explainability. Health-oriented AI research as a sub-field of digital health investigates a plethora of human-centered modalities. In this article, we address recent advances in the so far understudied but highly promising audio domain with a particular focus on speech data and present corresponding state-of-the-art technologies. Moreover, we give an excerpt of recent studies on the automatic audio-based detection of diseases ranging from acute and chronic respiratory diseases via psychiatric disorders to developmental disorders and neurodegenerative disorders. Our selection of presented literature shows that the recent success of deep learning methods in other fields of AI also more and more translates to the field of digital health, albeit expert-designed feature extractors and classical ML methodologies are still prominently used. Limiting factors, especially for speech-based disease detection systems, are related to the amount and diversity of available data, e. g., the number of patients and healthy controls as well as the underlying distribution of age, languages, and cultures. Finally, we contextualize and outline application scenarios of speech-based disease detection systems as supportive tools for health-care professionals under ethical consideration of privacy protection and faulty prediction.

Individuals with autism are known to face challenges with emotion regulation, and express their affective states in a variety of ways. With this in mind, an increasing amount of research on automatic affect recognition from speech and other modalities has recently been presented to assist and provide support, as well as to improve understanding of autistic individuals' behaviours. As well as the emotion expressed from the voice, for autistic children the dynamics of verbal speech can be inconsistent and vary greatly amongst individuals. The current contribution outlines a voice activity detection (VAD) system specifically adapted to autistic children's vocalisations. The presented VAD system is a recurrent neural network (RNN) with long short-term memory (LSTM) cells. It is trained on 130 acoustic Low-Level Descriptors (LLDs) extracted from more than 17 h of audio recordings, which were richly annotated by experts in terms of perceived emotion as well as occurrence and type of vocalisations. The data consist of 25 English-speaking autistic children undertaking a structured, partly robot-assisted emotion-training activity and was collected as part of the DE-ENIGMA project. The VAD system is further utilised as a preprocessing step for a continuous speech emotion recognition (SER) task aiming to minimise the effects of potential confounding information, such as noise, silence, or non-child vocalisation. Its impact on the SER performance is compared to the impact of other VAD systems, including a general VAD system trained from the same data set, an out-of-the-box Web Real-Time Communication (WebRTC) VAD system, as well as the expert annotations. Our experiments show that the child VAD system achieves a lower performance than our general VAD system, trained under identical conditions, as we obtain receiver operating characteristic area under the curve (ROC-AUC) metrics of 0.662 and 0.850, respectively. The SER results show varying performances across valence and arousal depending on the utilised VAD system with a maximum concordance correlation coefficient (CCC) of 0.263 and a minimum root mean square error (RMSE) of 0.107. Although the performance of the SER models is generally low, the child VAD system can lead to slightly improved results compared to other VAD systems and in particular the VAD-less baseline, supporting the hypothesised importance of child VAD systems in the discussed context.

Respiration is a key metabolic process in the marine environment and contemporary phytoplankton production (PhP) is commonly assumed the main driver. However, respiration in the absence of contemporary PhP, termed baseline respiration, can influence the energetics of an ecosystem and its sensitivity to hypoxia. Direct studies of baseline respiration are currently lacking. This study aims to obtain a first estimate of baseline respiration in a sub-arctic estuary and determine its contribution to plankton community respiration. Three approaches used to define baseline respiration determined the average rate to be 4.1 ± 0.1 (SE) mmol O2 m–3 d–1. A hypsographic model at the basin scale accounting for seasonal variation estimated an annual contribution of 30% baseline respiration to planktonic respiration. There was no correlation between plankton respiration and PhP, but a significant linear dependence was found with the total carbon supply from phytoplankton and riverine input. The sum of dissolved organic carbon transported by rivers, provided by both benthic and pelagic algae, could sustain 69% of the annual plankton respiration, of which as much as 25% occurred during winter. However, only 32% of the winter season respiration was explained, indicating that unknown carbon sources exist during the winter. Nitrification had a negligible (≤2.4%) effect on baseline respiration in the system. The results show that baseline respiration accounted for a significant percentage of coastal plankton respiration when allochthonous sources dominated the carbon supply, weakening the respiration-to- PhP relationship.

Spatiotemporal heterogeneity is one of the hallmarks of the relaxation dynamics associated with the glass transition. A key question in this regard is whether the dynamical heterogeneity has a structural origin. We report differential scanning calorimetry (DSC) data that reveal the presence of extreme spatial heterogeneity in the freezing dynamics of water (D2O) and the glass transition dynamics in the supercooled molecular liquid ortho-terphenyl (OTP), when these liquids are confined in nano-architected mesoporous silica with interconnected pores of different geometry. The results demonstrate, for the first time, that despite connectivity between the pores, nanoconfined water and OTP display distinct freezing/melting points and glass transition temperatures characteristic of each pore type. Viewed as a whole, these experimental results point to the coexistence of strong spatial heterogeneities over length scales of a few nanometers in the structure and dynamics of these liquids, suggesting a close mechanistic connection between them.

The impact of synoptic scale and mesoscale variability on the Lagrangian residence time (LRT) of the surface water in the Bay of Gdańsk was investigated using the results from an eddy-resolving model. The computed LRT of 53–60 days was up to four times longer than the estimated flushing time reported by Witek et al. (2003). The highest residence times were those of Puck Bay and near the coast, shallower than 50 m water depth, especially during the winter. These sites also had the highest annual mean in LRT. During the summer, when the level of biological activity is high, the LRT distribution was very heterogeneous and patchy, possibly due to the dynamics of varying eddy field and to variable wind forcing. Long-term run tracking of the inflowing water from the Vistula River (VR) showed a broad spectrum of tracer distribution. The potential impact of a much higher LRT on the near-coastal nitrogen cycle, coastal filter function and genetic differentiation is discussed, and the consequences for coastal zone management are considered. Since residence time is the most important factor regulating nutrient cycling, the incorporation of residence time into the Marine Strategy Framework Directive descriptors would result in an improved, unbiased evaluation of good environmental status.

Good sleep health is a key component to good personal well-being. It has been postulated that veterinary students have poor sleep health, but few measurements have been undertaken. This study measured Pittsburgh Sleep Quality Index and Epworth Sleepiness Scale values at multiple points throughout an academic year for students in a veterinary medical curriculum. Students were found to have overall poor sleep quality and above average to excessive daytime sleepiness. Further investigation is necessary to determine specific causes as well as what action can be taken to improve student sleep health.

The number of companion animal wellness visits in private practice has been decreasing, and one important factor cited is the lack of effective communication between veterinarians and pet owners regarding the importance of preventive care. Checklists have been widely used in many fields and are especially useful in areas where a complex task must be completed with multiple small steps, or when cognitive fatigue is evident. The use of checklists in veterinary medical education has not yet been thoroughly evaluated as a potential strategy to improve communication with pet owners regarding preventive care. The authors explored whether the use of a checklist based on the American Animal Hospital Association/American Veterinary Medical Association canine and feline preventive care guidelines would benefit senior veterinary students in accomplishing more complete canine and feline wellness visits. A group of students using provided checklists was compared to a control group of students who did not use checklists on the basis of their medical record notes from the visits. The students using the checklists were routinely more complete in several areas of a wellness visit vs. those who did not use the checklists. However, neither group of students routinely discussed follow-up care recommendations such as frequency or timing of follow-up visits. The study authors recommend considering checklist use for teaching and implementing wellness in companion animal primary care veterinary clinical teaching settings.

Among optogenetic tools, channelrhodopsins, the light gated ion channels of the plasma membrane from green algae, play the most important role. Properties like channel selectivity, timing parameters or color can be influenced by the exchange of selected amino acids. Although widely used, in the field of neurosciences for example, there is still little known about their photocycles and the mechanism of ion channel gating and conductance. One of the preferred methods for these studies is infrared spectroscopy since it allows observation of proteins and their function at a molecular level and in near-native environment. The absorption of a photon in channelrhodopsin leads to retinal isomerization within femtoseconds, the conductive states are reached in the microsecond time scale and the return into the fully dark-adapted state may take more than minutes. To be able to cover all these time regimes, a range of different spectroscopical approaches are necessary. This mini-review focuses on time-resolved applications of the infrared technique to study channelrhodopsins and other light triggered proteins. We will discuss the approaches with respect to their suitability to the investigation of channelrhodopsin and related proteins.

Recent years have seen a rapid increase in digital medicine research in an attempt to transform traditional healthcare systems to their modern, intelligent, and versatile equivalents that are adequately equipped to tackle contemporary challenges. This has led to a wave of applications that utilise AI technologies; first and foremost in the fields of medical imaging, but also in the use of wearables and other intelligent sensors. In comparison, computer audition can be seen to be lagging behind, at least in terms of commercial interest. Yet, audition has long been a staple assistant for medical practitioners, with the stethoscope being the quintessential sign of doctors around the world. Transforming this traditional technology with the use of AI entails a set of unique challenges. We categorise the advances needed in four key pillars: Hear, corresponding to the cornerstone technologies needed to analyse auditory signals in real-life conditions; Earlier, for the advances needed in computational and data efficiency; Attentively, for accounting to individual differences and handling the longitudinal nature of medical data; and, finally, Responsibly, for ensuring compliance to the ethical standards accorded to the field of medicine. Thus, we provide an overview and perspective of HEAR4Health: the sketch of a modern, ubiquitous sensing system that can bring computer audition on par with other AI technologies in the strive for improved healthcare systems.

Background and aimCurrent strategies for preventing diabetic sensorimotor polyneuropathy (DSPN) are limited mainly to glucose control but rapid decrease of glycemia can lead to acute onset or worsening of DSPN. The aim of this study was to examine the effects of periodic fasting on somatosensory nerve function in patients with type 2 diabetes (T2D).Study design and methodsSomatosensory nerve function was assessed in thirty-one patients with T2D (HbA1c 7.8 ± 1.3% [61.4 ± 14.3 mmol/mol]) before and after a six-month fasting-mimicking diet (FMD; n=14) or a control Mediterranean diet (M-diet; n=17). Neuropathy disability score (NDS), neuropathy symptoms score (NSS), nerve conduction velocity and quantitative sensory testing (QST) were analyzed. 6 participants of the M-Diet group and 7 of the FMD group underwent diffusion-weighted high-resolution magnetic resonance neurography (MRN) of the right leg before and after the diet intervention.ResultsClinical neuropathy scores did not differ between study groups at baseline (64% in the M-Diet group and 47% in the FMD group had DSPN) and no change was found after intervention. The differences in sensory NCV and sensory nerve action potential (SNAP) of sural nerve were comparable between study groups. Motor NCV of tibial nerve decreased by 12% in the M-Diet group (P=0.04), but did not change in the FMD group (P=0.39). Compound motor action potential (CMAP) of tibial nerve did not change in M-Diet group (P=0.8) and increased in the FMD group by 18% (P=0.02). Motor NCV and CMAP of peroneal nerve remained unchanged in both groups. In QST M-diet-group showed a decrease by 45% in heat pain threshold (P=0.02), FMD group showed no change (P=0.50). Changes in thermal detection, mechanical detection and mechanical pain did not differ between groups. MRN analysis showed stable fascicular nerve lesions irrespective of the degree of structural pathology. Fractional anisotropy and T2-time did not change in both study groups, while a correlation with the clinical degree of DSPN could be confirmed for both.ConclusionsOur study shows that six-month periodic fasting was safe in preserving nerve function and had no detrimental effects on somatosensory nerve function in T2D patients.Clinical trial registrationhttps://drks.de/search/en/trial/DRKS00014287, identifier DRKS00014287.

BackgroundDiabetic sensorimotor polyneuropathy (DSPN) is one of the most prevalent and poorly understood diabetic microvascular complications. Recent studies have found that fractional anisotropy (FA), a marker for microstructural nerve integrity, is a sensitive parameter for the structural and functional nerve damage in DSPN. The aim of this study was to investigate the significance of proximal sciatic nerve’s FA on different distal nerve fiber deficits of the upper and lower limbs and its correlation with the neuroaxonal biomarker, neurofilament light chain protein (NfL).Materials and methodsSixty-nine patients with type 2 diabetes (T2DM) and 30 healthy controls underwent detailed clinical and electrophysiological assessments, complete quantitative sensory testing (QST), and diffusion-weighted magnetic resonance neurography of the sciatic nerve. NfL was measured in the serum of healthy controls and patients with T2DM. Multivariate models were used to adjust for confounders of microvascular damage.ResultsPatients with DSPN showed a 17% lower sciatic microstructural integrity compared to healthy controls (p

In March 2020, Public Health England provided social distancing and shielding guidance for all adults aged 70 and over in response to the COVID-19 pandemic. This article seeks to provide insight into the lived experiences of older people during this period of household isolation. To do so, we analysed the metaphors used by 13 older adults during interviews discussing their experiences of household isolation, focusing on how these metaphors relate to a loss of agency. We found that participants negotiated their sense of agency through the use of metaphors involving physical force, movement, space, and animation of COVID-19. Metaphors were particularly used to discuss negative emotional impacts of the pandemic. Perceptions of a loss of agency were sometimes redressed through the use of comforting metaphors involving patterns and structure. In addition, participants explicitly rejected or refashioned dominant public metaphors that circulated as part of Government campaigns and wider public discourse to describe the pandemic and encourage certain behaviors. It has been argued that commonly used metaphors relating to containment, e.g., “bubble”, when applied to the context of household isolation, foreground the actions of those outside the container rather than those inside it, leading to a loss of feelings of agency. The participants' reactions to these suggest that common metaphors in public discourses are appropriated selectively and challenged by those at whom they are targeted. Hence, metaphor analysis can be used to paint a rich picture of the lived experience of older people experiencing household isolation, including their reaction to dominant public metaphors.

Virtual reality applications employing avatar embodiment typically use virtual mirrors to allow users to perceive their digital selves not only from a first-person but also from a holistic third-person perspective. However, due to distance-related biases such as the distance compression effect or a reduced relative rendering resolution, the self-observation distance (SOD) between the user and the virtual mirror might influence how users perceive their embodied avatar. Our article systematically investigates the effects of a short (1 m), middle (2.5 m), and far (4 m) SOD between users and mirror on the perception of their personalized and self-embodied avatars. The avatars were photorealistic reconstructed using state-of-the-art photogrammetric methods. Thirty participants repeatedly faced their real-time animated self-embodied avatars in each of the three SOD conditions, where they were repeatedly altered in their body weight, and participants rated the 1) sense of embodiment, 2) body weight perception, and 3) affective appraisal towards their avatar. We found that the different SODs are unlikely to influence any of our measures except for the perceived body weight estimation difficulty. Here, the participants perceived the difficulty significantly higher for the farthest SOD. We further found that the participants’ self-esteem significantly impacted their ability to modify their avatar’s body weight to their current body weight and that it positively correlated with the perceived attractiveness of the avatar. Additionally, the participants’ concerns about their body shape affected how eerie they perceived their avatars. The participants’ self-esteem and concerns about their body shape influenced the perceived body weight estimation difficulty. We conclude that the virtual mirror in embodiment scenarios can be freely placed and varied at a distance of one to four meters from the user without expecting major effects on the perception of the avatar.

With growing molecular evidence for correlations between spatial arrangement of blood vasculature and fundamental immunological functions, carried out in distinct compartments of the subdivided lymph node, there is an urgent need for three-dimensional models that can link these aspects. We reconstructed such models at a 1.84 µm resolution by the means of X-ray phase-contrast imaging with a 2D Talbot array in a short time without any staining. In addition reconstructions are verified in immunohistochemistry staining as well as in ultrastructural analyses. While conventional illustrations of mammalian lymph nodes depict the hilus as a definite point of blood and lymphatic vessel entry and exit, our method revealed that multiple branches enter and emerge from an area that extends up to one third of the organ’s surface. This could be a prerequisite for the drastic and location-dependent remodeling of vascularization, which is necessary for lymph node expansion during inflammation. Contrary to corrosion cast studies we identified B-cell follicles exhibiting a two times denser capillary network than the deep cortical units of the T-cell zone. In addition to our observation of high endothelial venules spatially surrounding the follicles, this suggests a direct connection between morphology and B-cell homing. Our findings will deepen the understanding of functional lymph node composition and lymphocyte migration on a fundamental basis.

Realistic and lifelike 3D-reconstruction of virtual humans has various exciting and important use cases. Our and others’ appearances have notable effects on ourselves and our interaction partners in virtual environments, e.g., on acceptance, preference, trust, believability, behavior (the Proteus effect), and more. Today, multiple approaches for the 3D-reconstruction of virtual humans exist. They significantly vary in terms of the degree of achievable realism, the technical complexities, and finally, the overall reconstruction costs involved. This article compares two 3D-reconstruction approaches with very different hardware requirements. The high-cost solution uses a typical complex and elaborated camera rig consisting of 94 digital single-lens reflex (DSLR) cameras. The recently developed low-cost solution uses a smartphone camera to create videos that capture multiple views of a person. Both methods use photogrammetric reconstruction and template fitting with the same template model and differ in their adaptation to the method-specific input material. Each method generates high-quality virtual humans ready to be processed, animated, and rendered by standard XR simulation and game engines such as Unreal or Unity. We compare the results of the two 3D-reconstruction methods in an immersive virtual environment against each other in a user study. Our results indicate that the virtual humans from the low-cost approach are perceived similarly to those from the high-cost approach regarding the perceived similarity to the original, human-likeness, beauty, and uncanniness, despite significant differences in the objectively measured quality. The perceived feeling of change of the own body was higher for the low-cost virtual humans. Quality differences were perceived more strongly for one’s own body than for other virtual humans.

The aim of this prospective observational trial was to evaluate the efficacy, toxicity and quality of life after stereotactic body radiation therapy (SBRT) in patients with hepatocellular carcinoma (HCC) and to assess the results of this treatment in comparison to trans-arterial chemoembolization (TACE). Patients with HCC, treated with TACE or SBRT, over a period of 12 months, enrolled in the study. The primary endpoint was feasibility; secondary endpoints were toxicity, quality of life (QOL), local progression (LP) and overall survival (OS). Between 06/2016 and 06/2017, 19 patients received TACE and 20 SBRT, 2 of whom were excluded due to progression. The median follow-up was 31 months. The QOL remained stable before and after treatment and was comparable in both treatment groups. Five patients developed grade ≥ 3 toxicities in the TACE group and 3 in the SBRT group. The cumulative incidence of LP after 1-, 2- and 3-years was 6, 6, 6% in the SBRT group and 28, 39, and 65% in the TACE group (p = 0.02). The 1- and 2- years OS rates were 84% and 47% in the TACE group and 44% and 39% in the SBRT group (p = 0.20). In conclusion, SBRT is a well-tolerated local treatment with a high local control rates and can be safely delivered, while preserving the QOL of HCC patients.