Deep learning-based voxel sampling for particle therapy treatment planning.

OBJECTIVE: Scanned particle therapy often requires complex treatment plans, robust optimization, as well as treatment adaptation. Plan optimization is especially complicated for heavy ions due to the variable relative biological effectiveness. We present a novel deep-learning model to select a subset of voxels in the planning process thus reducing the planning problem size for improved computational efficiency. Approach: Using only a subset of the voxels in target and organs at risk (OARs) we produced high-quality treatment plans, but heuristic selection strategies require manual input. We designed a deep-learning model based on P-Net to obtain an optimal voxel sampling without relying on patient-specific user input. A cohort of 70 head and neck patients that received carbon ion therapy was used for model training (50), validation (10) and testing (10). For training, a total of 12,500 carbon ion plans were optimized, using a highly efficient artificial intelligence (AI) infrastructure implemented into a research treatment planning platform. A custom loss function increased sampling density in underdosed regions, while aiming to reduce the total number of voxels. Main results: On the test dataset, the number of voxels in the optimization could be reduced by 84.8% (median) at <1% median loss in plan quality. When the model was trained to reduce sampling in the target only while keeping all voxels in OARs, a median reduction up to 71.6% was achieved, with 0.5% loss in the plan quality. The optimization time was reduced by a factor of 7.5 for the total AI selection model and a factor of 3.7 for the model with only target selection. Significance: The novel deep-learning voxel sampling technique achieves a significant reduction in computational time with a negligible loss in the plan quality. The reduction in optimization time can be especially useful for future real-time adaptation strategies. .

Video-based analysis of the blink reflex in Parkinson's disease patients.

We developed a video-based tool to quantitatively assess the Glabellar Tap Reflex (GTR) in patients with idiopathic Parkinson's disease (iPD) as well as healthy age-matched participants. We also video-graphically assessed the effect of dopaminergic medication on the GTR in iPD patients, as well as the frequency and blinking duration of reflex and non-reflex blinks. The Glabellar Tap Reflex is a clinical sign seen in patients e.g. suffering from iPD. Reliable tools to quantify this sign are lacking. METHODS: We recorded the GTR in 11 iPD patients and 12 healthy controls (HC) with a consumer-grade camera at a framerate of at least 180 images/s. In these videos, reflex and non-reflex blinks were analyzed for blink count and blinking duration in an automated fashion. RESULTS: With our setup, the GTR can be extracted from high-framerate cameras using landmarks of the MediaPipe face algorithm. iPD patients did not habituate to the GTR; dopaminergic medication did not alter that response. iPD patients' non-reflex blinks were higher in frequency and higher in blinking duration (width at half prominence); dopaminergic medication decreased the median frequency (Before medication-HC: p < 0.001, After medication-HC: p = 0.0026) and decreased the median blinking duration (Before medication-HC: p = 0.8594, After medication-HC: p = 0.6943)-both in the direction of HC. CONCLUSION: We developed a quantitative, video-based tool to assess the GTR and other blinking-specific parameters in HC and iPD patients. Further studies could compare the video data to electromyogram (EMG) data for accuracy and comparability, as well as evaluate the specificity of the GTR in patients with other neurodegenerative disorders, in whom the GTR can also be present. SIGNIFICANCE: The video-based detection of the blinking parameters allows for unobtrusive measurement in patients, a safer and more comfortable option.

Multiple instance learning framework can facilitate explainability in murmur detection.

OBJECTIVE: Cardiovascular diseases (CVDs) account for a high fatality rate worldwide. Heart murmurs can be detected from phonocardiograms (PCGs) and may indicate CVDs. Still, they are often overlooked as their detection and correct clinical interpretation require expert skills. In this work, we aim to predict the presence of murmurs and clinical outcomes from multiple PCG recordings employing an explainable multitask model. APPROACH: Our approach consists of a two-stage multitask model. In the first stage, we predict the murmur presence in single PCGs using a multiple instance learning (MIL) framework. MIL also allows us to derive sample-wise classifications (i.e. murmur locations) while only needing one annotation per recording ("weak label") during training. In the second stage, we fuse explainable hand-crafted features with features from a pooling-based artificial neural network (PANN) derived from the MIL framework. Finally, we predict the presence of murmurs and the clinical outcome for a single patient based on multiple recordings using a simple feed-forward neural network. MAIN RESULTS: We show qualitatively and quantitatively that the MIL approach yields useful features and can be used to detect murmurs on multiple time instances and may thus guide a practitioner through PCGs. We analyze the second stage of the model in terms of murmur classification and clinical outcome. We achieved a weighted accuracy of 0.714 and an outcome cost of 13612 when using the PANN model and demographic features on the CirCor dataset (hidden test set of the George B. Moody PhysioNet challenge 2022, team "Heart2Beat", rank 12 / 40). SIGNIFICANCE: To the best of our knowledge, we are the first to demonstrate the usefulness of MIL in PCG classification. Also, we showcase how the explainability of the model can be analyzed quantitatively, thus avoiding confirmation bias inherent to many post-hoc methods. Finally, our overall results demonstrate the merit of employing MIL combined with handcrafted features for the generation of explainable features as well as for a competitive classification performance.

An extensive quantitative analysis of the effects of errors in beat-to-beat intervals on all commonly used HRV parameters.

Heart rate variability (HRV) analysis is often used to estimate human health and fitness status. More specifically, a range of parameters that express the variability in beat-to-beat intervals are calculated from electrocardiogram beat detections. Since beat detection may yield erroneous interval data, these errors travel through the processing chain and may result in misleading parameter values that can lead to incorrect conclusions. In this study, we utilized Monte Carlo simulation on real data, Kolmogorov-Smirnov tests and Bland-Altman analysis to carry out extensive analysis of the noise sensitivity of different HRV parameters. The used noise models consider Gaussian and student-t distributed noise. As a result we observed that commonly used HRV parameters (e.g. pNN50 and LF/HF ratio) are especially sensitive to noise and that all parameters show biases to some extent. We conclude that researchers should be careful when reporting different HRV parameters, consider the distributions in addition to mean values, and consider reference data if applicable. The analysis of HRV parameter sensitivity to noise and resulting biases presented in this work generalizes over a wide population and can serve as a reference and thus provide a basis for the decision about which HRV parameters to choose under similar conditions.

Estimating Thoracic Movement with High-Sampling Rate THz Technology.

We use a high-sampling rate terahertz (THz) homodyne spectroscopy system to estimate thoracic movement from healthy subjects performing breathing at different frequencies. The THz system provides both the amplitude and phase of the THz wave. From the raw phase information, a motion signal is estimated. An electrocardiogram (ECG) signal is recorded with a polar chest strap to obtain ECG-derived respiration information. While the ECG showed sub-optimal performance for the purpose and only provided usable information for some subjects, the signal derived from the THz system showed good agreement with the measurement protocol. Over all the subjects, a root mean square estimation error of 1.40 BPM is obtained.

Multi-modal body part segmentation of infants using deep learning.

BACKGROUND: Monitoring the body temperature of premature infants is vital, as it allows optimal temperature control and may provide early warning signs for severe diseases such as sepsis. Thermography may be a non-contact and wireless alternative to state-of-the-art, cable-based methods. For monitoring use in clinical practice, automatic segmentation of the different body regions is necessary due to the movement of the infant. METHODS: This work presents and evaluates algorithms for automatic segmentation of infant body parts using deep learning methods. Based on a U-Net architecture, three neural networks were developed and compared. While the first two only used one imaging modality (visible light or thermography), the third applied a feature fusion of both. For training and evaluation, a dataset containing 600 visible light and 600 thermography images from 20 recordings of infants was created and manually labeled. In addition, we used transfer learning on publicly available datasets of adults in combination with data augmentation to improve the segmentation results. RESULTS: Individual optimization of the three deep learning models revealed that transfer learning and data augmentation improved segmentation regardless of the imaging modality. The fusion model achieved the best results during the final evaluation with a mean Intersection-over-Union (mIoU) of 0.85, closely followed by the RGB model. Only the thermography model achieved a lower accuracy (mIoU of 0.75). The results of the individual classes showed that all body parts were well-segmented, only the accuracy on the torso is inferior since the models struggle when only small areas of the skin are visible. CONCLUSION: The presented multi-modal neural networks represent a new approach to the problem of infant body segmentation with limited available data. Robust results were obtained by applying feature fusion, cross-modality transfer learning and classical augmentation strategies.

Video-Based Hand Movement Analysis of Parkinson Patients before and after Medication Using High-Frame-Rate Videos and MediaPipe.

Tremor is one of the common symptoms of Parkinson's disease (PD). Thanks to the recent evolution of digital technologies, monitoring of PD patients' hand movements employing contactless methods gained momentum. Objective: We aimed to quantitatively assess hand movements in patients suffering from PD using the artificial intelligence (AI)-based hand-tracking technologies of MediaPipe. Method: High-frame-rate videos and accelerometer data were recorded from 11 PD patients, two of whom showed classical Parkinsonian-type tremor. In the OFF-state and 30 Minutes after taking their standard oral medication (ON-state), video recordings were obtained. First, we investigated the frequency and amplitude relationship between the video and accelerometer data. Then, we focused on quantifying the effect of taking standard oral treatments. Results: The data extracted from the video correlated well with the accelerometer-based measurement system. Our video-based approach identified the tremor frequency with a small error rate (mean absolute error 0.229 (±0.174) Hz) and an amplitude with a high correlation. The frequency and amplitude of the hand movement before and after medication in PD patients undergoing medication differ. PD Patients experienced a decrease in the mean value for frequency from 2.012 (±1.385) Hz to 1.526 (±1.007) Hz and in the mean value for amplitude from 8.167 (±15.687) a.u. to 4.033 (±5.671) a.u. Conclusions: Our work achieved an automatic estimation of the movement frequency, including the tremor frequency with a low error rate, and to the best of our knowledge, this is the first paper that presents automated tremor analysis before/after medication in PD, in particular using high-frame-rate video data.

Accuracy enhancement in reflective pulse oximetry by considering wavelength-dependent pathlengths.

Objective.Noninvasive measurement of oxygen saturation (SpO2) using transmissive photoplethysmography (tPPG) is clinically accepted and widely employed. However, reflective photoplethysmography (rPPG)-currently present in smartwatches-has not become equally accepted, partially because the pathlengths of the red and infrared PPGs are patient-dependent. Thus, even the most popular 'Ratio of Modulation' (R) method requires patient-dependent calibration to reduce the errors in the measurement ofSpO2using rPPGs.Approach.In this paper, a correction factor or 'pathlength ratio'ßis introduced in an existing calibration-free algorithm that compensates the patient-dependent pathlength variations, and improved accuracy is obtained in the measurement ofSpO2using rPPGs. The proposed pathlength ratioßis derived through the analytical model of a rPPG signal. Using the new expression and data obtained from a human hypoxia study wherein arterial oxygen saturation values acquired through Blood Gas Analysis were employed as a reference,ßis determined.Main results.The results of the analysis show that a specific combination of theßand the measurements on the pulsating part of the natural logarithm of the red and infrared PPG signals yields a reduced root-mean-square error (RMSE). It is shown that the average RMSE in measuringSpO2values reduces to 1 %.Significance.The human hypoxia study data used for this work, obtained in a previous study, coversSpO2values in the range from 70 % to 100 %, and thus shows that the pathlength ratioßproposed here works well in the range of clinical interest. This work demonstrates that the calibration-free method applicable for transmission type PPGs can be extended to determineSpO2using reflective PPGs with the incorporation of the correction factorß. Our algorithm significantly reduces the number of parameters needed for the estimation, while keeping the RMSE below the clinically accepted 2 %.

Skin tissue perfusion mapping triggered by an audio-(de)modulated reference signal.

Spatial mapping of skin perfusion provides essential information about physiological processes that are often hidden from the eyes of the examining physician. The perfusion map quality depends on several key factors, such as the camera system type, frame rate, sensitivity, or signal-to-noise ratio. When investigating physiological parameters, the reference signal allows for increasing the spatial resolution of the photoplethysmography imaging (PPGI) system. On the other hand, it increases the system complexity and the synchronization prerequisites. Our solution is a hardware device that modulates the reference biosignal into the audio frequency band. This signal is connected to the mic input of a digital camera or a smartphone, enabling the transformation of such a device into a PPGI measurement system even in the case of compressed video recording using lock-in amplification technique. It also brings the possibility of synchronous recording of PPGI and another reference signal such as conventional photoplethysmogram or electrocardiogram.

Estimation of the respiratory rate from ballistocardiograms using the Hilbert transform.

BACKGROUND: Measuring the respiratory rate is usually associated with discomfort for the patient due to contact sensors or a high time demand for healthcare personnel manually counting it. METHODS: In this paper, two methods for the continuous extraction of the respiratory rate from unobtrusive ballistocardiography signals are introduced. The Hilbert transform is used to generate an amplitude-invariant phase signal in-line with the respiratory rate. The respiratory rate can then be estimated, first, by using a simple peak detection, and second, by differentiation. RESULTS: By analysis of a sleep laboratory data set consisting of nine records of healthy individuals lasting more than 63 h and including more than 59,000 breaths, a mean absolute error of as low as 0.7 BPM for both methods was achieved. CONCLUSION: The results encourage further assessment for hospitalised patients and for home-care applications especially with patients suffering from diseases of the respiratory system like COPD or sleep apnoea.

Exploring novel algorithms for atrial fibrillation detection by driving graduate level education in medical machine learning.

During the lockdown of universities and the COVID-Pandemic most students were restricted to their homes. Novel and instigating teaching methods were required to improve the learning experience and so recent implementations of the annual PhysioNet/Computing in Cardiology (CinC) Challenges posed as a reference. For over 20 years, the challenges have proven repeatedly to be of immense educational value, besides leading to technological advances for specific problems. In this paper, we report results from the class 'Artificial Intelligence in Medicine Challenge', which was implemented as an online project seminar at Technical University Darmstadt, Germany, and which was heavily inspired by the PhysioNet/CinC Challenge 2017 'AF Classification from a Short Single Lead ECG Recording'. Atrial fibrillation is a common cardiac disease and often remains undetected. Therefore, we selected the two most promising models of the course and give an insight into the Transformer-based DualNet architecture as well as into the CNN-LSTM-based model and finally a detailed analysis for both. In particular, we show the model performance results of our internal scoring process for all submitted models and the near state-of-the-art model performance for the two named models on the official 2017 challenge test set. Several teams were able to achieve F1scores above/close to 90% on a hidden test-set of Holter recordings. We highlight themes commonly observed among participants, and report the results from the self-assessed student evaluation. Finally, the self-assessment of the students reported a notable increase in machine learning knowledge.

Camera fusion for real-time temperature monitoring of neonates using deep learning.

The continuous monitoring of vital signs is a crucial aspect of medical care in neonatal intensive care units. Since cable-based sensors pose a potential risk for the immature skin of preterm infants, unobtrusive monitoring techniques using camera systems are increasingly investigated. The combination of deep learning-based algorithms and camera modalities such as RGB and infrared thermography can improve the development of cable-free methods for the extraction of vital parameters. In this study, a real-time approach for local extraction of temperatures on the body surface of neonates using a multi-modal clinical dataset was implemented. Therefore, a trained deep learning-based keypoint detector was used for body landmark prediction in RGB. Image registration was conducted to transfer the RGB points to the corresponding thermographic recordings. These landmarks were used to extract the body surface temperature in various regions to determine the central-peripheral temperature difference. A validation of the keypoint detector showed a mean average precision of 0.82. The registration resulted in mean absolute errors of 16.4 px (8.2 mm) for x and 22.4 px (11.2 mm) for y. The evaluation of the temperature extraction revealed a mean absolute error of 0.55 [Formula: see text]C. A final performance of 31 fps was observed on the NVIDIA Jetson Xavier NX module, which proves real-time capability on an embedded GPU system. As a result, the approach can perform real-time temperature extraction on a low-cost GPU module.

Non-Contact Measurement of Heart Rate Variability in Frail Geriatric Patients: Response to Early Geriatric Rehabilitation and Comparison with Healthy Old Community-Dwelling Individuals - A Pilot Study.

INTRODUCTION: Frailty is a central geriatric syndrome characterized by a state of increased physiological vulnerability. As the key components of frailty are difficult to capture in their entirety, easily measurable and reliable surrogate parameters are desirable. Since frailty influences heart rate variability (HRV), HRV may be such a surrogate parameter. HRV is typically acquired by an ECG, which, however, may not be tolerated by all patients; in some, it may even trigger delirium. Therefore, we sought to measure HRV in a non-contact and unobtrusive way through photoplethysmography imaging (PPGI). Using our previously presented HRV estimation algorithm for PPGI, we investigated whether PPGI could reveal (1) HRV differences between frail and non-frail individuals and (2) the influences of early geriatric rehabilitation on HRV. METHODS: The study involved 10 frail geriatric inpatients undergoing early geriatric rehabilitation and 10 healthy community-dwelling older adults. All participants underwent a comprehensive geriatric assessment. HRV measurements using a PPGI system and a reference ECG were made at the beginning and the end of the rehabilitation. HRV in terms of LF/HF ratio was analysed for both intra-individual changes during the geriatric rehabilitation and differences between frail geriatric patients and healthy community-dwelling individuals. RESULTS: Across all geriatric patients, the median LF/HF ratio obtained with PPGI was found to be reduced by 0.178 (24.8%) during early geriatric rehabilitation. The assessment at the end of the rehabilitation revealed a simultaneous improvement of the functional state. Moreover, frail geriatric patients had a higher LF/HF ratio than their community-dwelling counterparts. Both observations in PPGI-based HRV were confirmed by the reference. The capability of PPGI to track intra-individual HRV changes was also analysed; a Spearman correlation of ρ = 1.0 between PPGI-based HRV and reference was achieved for 58.8% of the participants. CONCLUSION: Early geriatric rehabilitation improves the functional state, which is associated with an increased HRV. PPGI is capable of detecting HRV changes/trends in that age group. While the tracking of intra-individual HRV changes is also possible, its reliability needs improvement. Nevertheless, the capabilities demonstrated in our study and the non-contact measurement principle of PPGI emphasize its potential for application in geriatric medicine.

Acceleration of Longitudinal Track and Field Performance Declines in Athletes Who Still Compete at the Age of 100 Years.

While physical performance decline rates accelerate after around the age of 70 years, longitudinal athletic performance trends in athletes older than 95 years are unknown. We hypothesized a further accelerated decline in human performance in athletes who still perform at the age of 100 years. To investigate this, longitudinal data of all athletes with results at or over the age of 100 years were collected from the "World Master Rankings" data base spanning 2006-2019 (138 results from 42 athletes; 5 women, 37 men; maximum 105 years) and compared to previously published longitudinal data from 80- to 96-year-old athletes from Sweden (1,134 results from 374 athletes). Regression statistics were used to compare performance decline rates between disciplines and age groups. On average, the individual decline rate of the centenarian group was 2.53 times as steep (100 m: 8.22x; long jump: 0.82x; shot put: 1.61x; discus throw: 1.04x; javelin throw: 0.98x) as that seen in non-centenarians. The steepest increase in decline was found in the 100-m sprint (t-test: p < 0.05, no sign. difference in the other disciplines). The pooled regression statistics of the centenarians are: 100 m: R = 0.57, p = 0.004; long jump: R = 0.90, p < 0.001; shot put: R = 0.65, p < 0.001; discus throw: R = 0.73, p < 0.001; javelin throw: R = 0.68, p < 0.001. This first longitudinal dataset of performance decline rates of athletes who still compete at 100 years and older in five athletics disciplines shows that there is no performance plateau after the age of 90, but rather a further acceleration of the performance decline.

Learning from machine learning: prediction of age-related athletic performance decline trajectories.

Factors that determine individual age-related decline rates in physical performance are poorly understood and prediction poses a challenge. Linear and quadratic regression models are usually applied, but often show high prediction errors for individual athletes. Machine learning approaches may deliver more accurate predictions and help to identify factors that determine performance decline rates. We hypothesized that it is possible to predict the performance development of a master athlete from a single measurement, that prediction by a machine learning approach is superior to prediction by the average decline curve or an individually shifted decline curve, and that athletes with a higher starting performance show a slower performance decline than those with a lower performance. The machine learning approach was implemented using a multilayer neuronal network. Results showed that performance prediction from a single measurement is possible and that the prediction by a machine learning approach was superior to the other models. The estimated performance decline rate was highest in athletes with a high starting performance and a low starting age, as well as in those with a low starting performance and high starting age, while the lowest decline rate was found for athletes with a high starting performance and a high starting age. Machine learning was superior and predicted trajectories with significantly lower prediction errors compared to conventional approaches. New insights into factors determining decline trajectories were identified by visualization of the model outputs. Machine learning models may be useful in revealing unknown factors that determine the age-related performance decline.

A model-based source separation algorithm for lung perfusion imaging using electrical impedance tomography.

Objective. Electrical impedance tomography (EIT) for lung perfusion imaging is attracting considerable interest in intensive care, as it might open up entirely new ways to adjust ventilation therapy. A promising technique is bolus injection of a conductive indicator to the central venous catheter, which yields the indicator-based signal (IBS). Lung perfusion images are then typically obtained from the IBS using the maximum slope technique. However, the low spatial resolution of EIT results in a partial volume effect (PVE), which requires further processing to avoid regional bias.Approach. In this work, we repose the extraction of lung perfusion images from the IBS as a source separation problem to account for the PVE. We then propose a model-based algorithm, called gamma decomposition (GD), to derive an efficient solution. The GD algorithm uses a signal model to transform the IBS into a parameter space where the source signals of heart and lung are separable by clustering in space and time. Subsequently, it reconstructs lung model signals from which lung perfusion images are unambiguously extracted.Main results. We evaluate the GD algorithm on EIT data of a prospective animal trial with eight pigs. The results show that it enables lung perfusion imaging using EIT at different stages of regional impairment. Furthermore, parameters of the source signals seem to represent physiological properties of the cardio-pulmonary system.Significance. This work represents an important advance in IBS processing that will likely reduce bias of EIT perfusion images and thus eventually enable imaging of regional ventilation/perfusion (V/Q) ratio.

Accuracy of heart rate variability estimated with reflective wrist-PPG in elderly vascular patients.

Optical heart rate monitoring (OHR) with reflective wrist photoplethysmography is a technique mainly used in the wellness application domain for monitoring heart rate levels during exercise. In the absence of motion, OHR technique is also able to estimate individual beat-to-beat intervals relatively well and can therefore also be used, for example, in monitoring of cardiac arrhythmias, stress, or sleep quality through heart rate variability (HRV) analysis. HRV analysis has also potential in monitoring the recovery of patients, e.g. after a medical intervention. However, in order to detect subtle changes, the calculated HRV parameters should be sufficiently accurate and very few studies exist that asses the accuracy of OHR derived HRV in non-healthy subjects. In this paper, we present a method to estimate beat-to-beat-intervals (BBIs) from reflective wrist PPG signal and evaluated the accuracy of the proposed method in estimating BBIs in a cross-sectional study with 29 hospitalized patients (mean age 70.6 years) in 24-h recordings performed after peripheral vascular surgery or endovascular interventions. Finally, we evaluate the accuracy of more than 30 commonly used HRV parameters and find that the accuracy of certain metrics, for example SDNN and triangular index, shown in the literature to be associated with the deterioration of the status of the patients during recovery from surgical intervention, could be adequate for patient monitoring. On the other hand, the parameters more affected by the high-frequency content of the HRV and especially the LF/HF-ratio should be used with caution.

YouTube Videos on Parkinson's Disease are a Relevant Source of Patient Information.

BACKGROUND: Parkinson's disease (PD) is the most frequent movement disorder. Patients access YouTube, one of the largest video databases in the world, to retrieve health-related information increasingly often. OBJECTIVE: We aimed to identify high-quality publishers, so-called "channels" that can be recommended to patients. We hypothesized that the number of views and the number of uploaded videos were indicators for the quality of the information given by a video on PD. METHODS: YouTube was searched for 8 combinations of search terms that included "Parkinson" in German. For each term, the first 100 search results were analyzed for source, date of upload, number of views, numbers of likes and dislikes, and comments. The view ratio (views / day) and the likes ratio (likes * 100 / [likes + dislikes]) were determined to calculate the video popularity index (VPI). The global quality score (GQS) and title - content consistency index (TCCI) were assessed in a subset of videos. RESULTS: Of 800 search results, 251 videos met the inclusion criteria. The number of views or the publisher category were not indicative of higher quality video content. The number of videos uploaded by a channel was the best indicator for the quality of video content. CONCLUSION: The quality of YouTube videos relevant for PD patients is increased in channels with a high number of videos on the topic. We identified three German channels that can be recommended to PD patients who prefer video over written content.

A Deep Learning-Based Camera Approach for Vital Sign Monitoring Using Thermography Images for ICU Patients.

Infrared thermography for camera-based skin temperature measurement is increasingly used in medical practice, e.g., to detect fevers and infections, such as recently in the COVID-19 pandemic. This contactless method is a promising technology to continuously monitor the vital signs of patients in clinical environments. In this study, we investigated both skin temperature trend measurement and the extraction of respiration-related chest movements to determine the respiratory rate using low-cost hardware in combination with advanced algorithms. In addition, the frequency of medical examinations or visits to the patients was extracted. We implemented a deep learning-based algorithm for real-time vital sign extraction from thermography images. A clinical trial was conducted to record data from patients on an intensive care unit. The YOLOv4-Tiny object detector was applied to extract image regions containing vital signs (head and chest). The infrared frames were manually labeled for evaluation. Validation was performed on a hold-out test dataset of 6 patients and revealed good detector performance (0.75 intersection over union, 0.94 mean average precision). An optical flow algorithm was used to extract the respiratory rate from the chest region. The results show a mean absolute error of 2.69 bpm. We observed a computational performance of 47 fps on an NVIDIA Jetson Xavier NX module for YOLOv4-Tiny, which proves real-time capability on an embedded GPU system. In conclusion, the proposed method can perform real-time vital sign extraction on a low-cost system-on-module and may thus be a useful method for future contactless vital sign measurements.

Longitudinal Master Track and Field Performance Decline Rates Are Lower and Performance Is Better Compared to Athletes Competing Only Once.

In master athletics research, cross-sectional data are easier to obtain than longitudinal data. While cross-sectional data give the age-related performance decline for a population, longitudinal data show individual trajectories. It is not known whether athletes who repeatedly compete have (a) a better performance and (b) a slower age-related decline in performance than that obtained from cross-sectional data from athletes competing only once. To investigate this, we analyzed 33 254 results of 14 118 male athletes from 8 disciplines in the database of "Swedish Veteran Athletics." For each discipline and for the pooled data of all disciplines, quadratic models of the evolution of performance over time were analyzed by ANCOVA/ANOCOVA using MATLAB. The performance was higher in athletes with 2 or more data points compared to those with only n = 1 (p < .001), with further increases in performance with an increasing number of data points per athlete. The estimated performance decline was lower in people with 2 or more results (sprint, 10 km, jumps; p < .001). In conclusion, we showed that longitudinal data are associated with higher performance and lower performance decline rates.