Evaluating the risk of endometriosis based on patients' self-assessment questionnaires.

BACKGROUND: Endometriosis is a condition that significantly affects the quality of life of about 10 % of reproductive-aged women. It is characterized by the presence of tissue similar to the uterine lining (endometrium) outside the uterus, which can lead lead scarring, adhesions, pain, and fertility issues. While numerous factors associated with endometriosis are documented, a wide range of symptoms may still be undiscovered. METHODS: In this study, we employed machine learning algorithms to predict endometriosis based on the patient symptoms extracted from 13,933 questionnaires. We compared the results of feature selection obtained from various algorithms (i.e., Boruta algorithm, Recursive Feature Selection) with experts' decisions. As a benchmark model architecture, we utilized a LightGBM algorithm, along with Multivariate Imputation by Chained Equations (MICE) and k-nearest neighbors (KNN), for missing data imputation. Our primary objective was to assess the model's performance and feature importance compared to existing studies. RESULTS: We identified the top 20 predictors of endometriosis, uncovering previously overlooked features such as Cesarean section, ovarian cysts, and hernia. Notably, the model's performance metrics were maximized when utilizing a combination of multiple feature selection methods. Specifically, the final model achieved an area under the receiver operator characteristic curve (AUC) of 0.85 on the training dataset and an AUC of 0.82 on the testing dataset. CONCLUSIONS: The application of machine learning in diagnosing endometriosis has the potential to significantly impact clinical practice, streamlining the diagnostic process and enhancing efficiency. Our questionnaire-based prediction approach empowers individuals with endometriosis to proactively identify potential symptoms, facilitating informed discussions with healthcare professionals about diagnosis and treatment options.

Thermal Image Processing for Respiratory Estimation from Cubical Data with Expandable Depth.

As healthcare costs continue to rise, finding affordable and non-invasive ways to monitor vital signs is increasingly important. One of the key metrics for assessing overall health and identifying potential issues early on is respiratory rate (RR). Most of the existing methods require multiple steps that consist of image and signal processing. This might be difficult to deploy on edge devices that often do not have specialized digital signal processors (DSP). Therefore, the goal of this study is to develop a single neural network realizing the entire process of RR estimation in a single forward pass. The proposed solution builds on recent advances in video recognition, capturing both spatial and temporal information in a multi-path network. Both paths process the data at different sampling rates to capture rapid and slow changes that are associated with differences in the temperature of the nostril area during the breathing episodes. The preliminary results show that the introduced end-to-end solution achieves better performance compared to state-of-the-art methods, without requiring additional pre/post-processing steps and signal-processing techniques. In addition, the presented results demonstrate its robustness on low-resolution thermal video sequences that are often used at the embedded edge due to the size and power constraints of such systems. Taking that into account, the proposed approach has the potential for efficient and convenient respiratory rate estimation across various markets in solutions deployed locally, close to end users.

Mild poikilocapnic hypoxia increases very low frequency haemoglobin oxygenation oscillations in prefrontal cortex.

BACKGROUND: The aim of the study was to investigate the effect of mild cerebral hypoxia on haemoglobin oxygenation (HbO2), cerebrospinal fluid dynamics and cardiovascular physiology. To achieve this goal, four signals were recorded simultaneously: blood pressure, heart rate / electrocardiogram, HbO2 from right hemisphere and changes of subarachnoid space (SAS) width from left hemisphere. Signals were registered from 30 healthy, young participants (2 females and 28 males, body mass index = 24.5 ± 2.3 kg/m2, age 30.8 ± 13.4 years). RESULTS: We analysed the recorded signals using wavelet transform and phase coherence. We demonstrated for the first time that in healthy subjects exposed to mild poikilokapnic hypoxia there were increases in very low frequency HbO2 oscillations (< 0.052 Hz) in prefrontal cortex. Additionally, SAS fluctuation diminished in the whole frequency range which could be explained by brain oedema. CONCLUSIONS: Consequently the study provides insight into mechanisms governing brain response to a mild hypoxic challenge. Our study supports the notion that HbO2 and SAS width monitoring might be beneficial for patients with acute lung disease.

Face with Mask Detection in Thermal Images Using Deep Neural Networks.

As the interest in facial detection grows, especially during a pandemic, solutions are sought that will be effective and bring more benefits. This is the case with the use of thermal imaging, which is resistant to environmental factors and makes it possible, for example, to determine the temperature based on the detected face, which brings new perspectives and opportunities to use such an approach for health control purposes. The goal of this work is to analyze the effectiveness of deep-learning-based face detection algorithms applied to thermal images, especially for faces covered by virus protective face masks. As part of this work, a set of thermal images was prepared containing over 7900 images of faces with and without masks. Selected raw data preprocessing methods were also investigated to analyze their influence on the face detection results. It was shown that the use of transfer learning based on features learned from visible light images results in mAP greater than 82% for half of the investigated models. The best model turned out to be the one based on Yolov3 model (mean average precision-mAP, was at least 99.3%, while the precision was at least 66.1%). Inference time of the models selected for evaluation on a small and cheap platform allows them to be used for many applications, especially in apps that promote public health.

Mild poikilocapnic hypoxia increases very low frequency haemoglobin oxygenation oscillations in prefrontal cortex

BACKGROUND: The aim of the study was to investigate the effect of mild cerebral hypoxia on haemoglobin oxygenation (HbO2), cerebrospinal fluid dynamics and cardiovascular physiology. To achieve this goal, four signals were recorded simultaneously: blood pressure, heart rate / electrocardiogram, HbO2 from right hemisphere and changes of subarachnoid space (SAS) width from left hemisphere. Signals were registered from 30 healthy, young participants (2 females and 28 males, body mass index = 24.5 ± 2.3 kg/m2, age 30.8 ± 13.4 years). RESULTS: We analysed the recorded signals using wavelet transform and phase coherence. We demonstrated for the first time that in healthy subjects exposed to mild poikilokapnic hypoxia there were increases in very low frequency HbO2 oscillations (< 0.052 Hz) in prefrontal cortex. Additionally, SAS fluctuation diminished in the whole frequency range which could be explained by brain oedema. CONCLUSIONS: Consequently the study provides insight into mechanisms governing brain response to a mild hypoxic challenge. Our study supports the notion that HbO2 and SAS width monitoring might be beneficial for patients with acute lung disease.

Comparison of near infrared spectroscopy (NIRS) and near-infrared transillumination-backscattering sounding (NIR-T/BSS) methods.

The aim of the study was to compare simultaneously recorded a NIR-T/BSS and NIRS signals from healthy volunteers. NIR-T/BSS is a device which give an ability to non-invasively detect and monitor changes in the subarachnoid space width (SAS). Experiments were performed on a group of 30 healthy volunteers (28 males and 2 females, age 30.8 ± 13.4 years, BMI = 24.5 ± 2.3 kg/m2). We analysed recorded signals using analysis methods based on wavelet transform (WT) for the wide frequency range from 0.0095 to 2 Hz. Despite the fact that both devices use a similar radiation source both signals are distinct from each other. We found statistically significant differences for WT amplitude spectra between both signals. Additionally, we showed different relationships of both signals to blood pressure. Collectively, based on the present findings and those of previous studies, we can conclude that the combination of NIR-T/BSS or NIRS signals and time-frequency analysis opens new frontiers in science, and give possibility to understand and diagnosis of various neurodegenerative and ageing related diseases to improve diagnostic procedures and patient prognosis.

Evaluating Accuracy of Respiratory Rate Estimation from Super Resolved Thermal Imagery.

Non-contact estimation of Respiratory Rate (RR) has revolutionized the process of establishing the measurement by surpassing some issues related to attaching sensors to a body, e.g. epidermal stripping, skin disruption and pain. In this study, we perform further experiments with image processing-based RR estimation by using various image enhancement algorithms. Specifically, we employ Super Resolution (SR) Deep Learning (DL) network to generate hallucinated thermal image sequences that are then analyzed to extract breathing signals. DL-based SR networks have been proved to increase image quality in terms of Peak Signal-to-Noise ratio. However, it hasn't been evaluated yet whether it leads to better RR estimation accuracy, what we address in this study. Our research confirms that for estimator based on the dominated peak in the frequency spectrum Root Mean Squared Error improves by 0.15bpm for 8-bit and by 0.84bpm for 16-bit data comparing to original sequences if hallucinated frames are used. Mean Absolute Error is reduced by 0.63bpm for average aggregator and by 2.06bpm for skewness. This finding can enable various remote monitoring solutions that may suffer from poorer accuracy due to low spatial resolution of utilized thermal cameras.

Evaluation of Facial Pulse Signals using Deep Neural Net Models.

The reliable measurement of the pulse rate using remote photoplethysmography (PPG) is very important for many medical applications. In this paper we present how deep neural networks (DNNs) models can be used in the problem of PPG signal classification and pulse rate estimation. In particular, we show that the DNN-based classification results correspond to parameters describing the PPG signals (e.g. peak energy in the frequency domain, SNR, etc.). The results show that it is possible to identify regions of a face, for which reliable PPG signals can be extracted. The accuracy obtained for the classification task and the mean absolute error achieved for the regression task proved the usefulness of the DNN models.

A Meta-Analysis of Pulse Arrival Time Based Blood Pressure Estimation.

The paper presents a preliminary meta-analysis of the sample correlation between pulse arrival time (PAT) and blood pressure (BP). The aim of the study was to verify sample correlation coefficient between PAT and BP using an affine model $ BP = a . PAT+b $ for systolic and diastolic blood pressure. The databases included in the search were the IEEE Xplore Digital Library, Springer Link and Google Scholar. Only papers from 2005 to 2017 wereincluded into analysis. The random-effects model was considered. The resulting sample correlation coefficient was equal to -0.82 (95 % CI; -0.89, -0.72) for systolic blood pressure and -0.64 (95% CI, -0.74 -0.51) for diastolic one. Egger's regression test showed that there was no evidence of publication bias. Obtained 95% CI intervals for sample correlation coefficients for SBP and DBP are almost separate, which may indicate different relation between PAT and BP for systolic and diastolic pressure.

Analysis of the parameters of respiration patterns extracted from thermal image sequences.

Remote estimation of vital signs is an important and active area of research. The goal of this work was to analyze the feasibility of estimating respiration parameters from video sequences of faces recorded using a mobile thermal camera. Different estimators were analyzed and experimentally verified. It was demonstrated that the respiration rate, periodicity of respiration, and presence and length of apnea periods could be reliably estimated from signals recorded using a portable thermal camera. The size of the camera and efficiency of the methods allow the implementation of this method in smart glasses.

Application of smart glasses for fast and automatic color correction in health care.

In recent years different applications of smart glasses in health care have been proposed. In this paper we present the experiments related to automatic color correction using smart glasses platform developed within the eGlasses project. The color pattern is proposed and tested enabling the automatic detection of the pattern and automatic correction of colors. Additionally, the method for encoding and decoding of patient ID in the color pattern is presented. This enables automatic data integration using smart glasses connected to Hospital Information System or similar systems.

Application of adjusted subpixel method (ASM) in HRCT measurements of the bronchi in bronchial asthma patients and healthy individuals.

BACKGROUND: Recently, we described a model system which included corrections of high-resolution computed tomography (HRCT) bronchial measurements based on the adjusted subpixel method (ASM). OBJECTIVE: To verify the clinical application of ASM by comparing bronchial measurements obtained by means of the traditional eye-driven method, subpixel method alone and ASM in a group comprised of bronchial asthma patients and healthy individuals. METHODS: The study included 30 bronchial asthma patients and the control group comprised of 20 volunteers with no symptoms of asthma. The lowest internal and external diameters of the bronchial cross-sections (ID and ED) and their derivative parameters were determined in HRCT scans using: (1) traditional eye-driven method, (2) subpixel technique, and (3) ASM. RESULTS: In the case of the eye-driven method, lower ID values along with lower bronchial lumen area and its percentage ratio to total bronchial area were basic parameters that differed between asthma patients and healthy controls. In the case of the subpixel method and ASM, both groups were not significantly different in terms of ID. Significant differences were observed in values of ED and total bronchial area with both parameters being significantly higher in asthma patients. Compared to ASM, the eye-driven method overstated the values of ID and ED by about 30% and 10% respectively, while understating bronchial wall thickness by about 18%. CONCLUSIONS: Results obtained in this study suggest that the traditional eye-driven method of HRCT-based measurement of bronchial tree components probably overstates the degree of bronchial patency in asthma patients.

Security of distributed processing of medical image data using JINI technology.

Health Care Organizations (HCOs) are highly computerized using broadband networks for data communication within and outside of their departments. However their computer resources are usually task oriented, and not effectively used. Already introduced and new standards for e-Health (e.g. DICOM, TC251-standards) offer workflow management models to organize different resources towards a one, complex task. Distributed processing can be used to implement such models. Similarly, computational resources can be used for any complicated problem-solving task inside a HCO department, between HCO departments and between HCOs. In this article the security analysis of Jini-based computational grids and cost of safeguards introduction is presented. The Jini-based GRID was analyzed in two medical parametric imaging tasks: active, dynamic infrared imaging and dynamic susceptibility contrast imaging (DSC-MRI).

Representation of thermal infrared imaging data in the DICOM using XML configuration files.

The DICOM standard has become a widely accepted and implemented format for the exchange and storage of medical imaging data. Different imaging modalities are supported however there is not a dedicated solution for thermal infrared imaging in medicine. In this article we propose new ideas and improvements to final proposal of the new DICOM Thermal Infrared Imaging structures and services. Additionally, we designed, implemented and tested software packages for universal conversion of existing thermal imaging files to the DICOM format using XML configuration files. The proposed solution works fast and requires minimal number of user interactions. The XML configuration file enables to compose a set of attributes for any source file format of thermal imaging camera.

Thermal parametric imaging in the evaluation of skin burn depth.

The aim of this paper is to determine the extent to which infrared (IR) thermal imaging may be used for skin burn depth evaluation. The analysis can be made on the basis of the development of a thermal model of the burned skin. Different methods such as the traditional clinical visual approach and the IR imaging modalities of static IR thermal imaging, active IR thermal imaging and active-dynamic IR thermal imaging (ADT) are analyzed from the point of view of skin burn depth diagnostics. In ADT, a new approach is proposed on the basis of parametric image synthesis. Calculation software is implemented for single-node and distributed systems. The properties of all the methods are verified in experiments using phantoms and subsequently in vivo with animals with a reference histopathological examination. The results indicate that it is possible to distinguish objectively and quantitatively burns which will heal spontaneously within three weeks of infliction and which should be treated conservatively from those which need surgery because they will not heal within this period.

Burn depths evaluation based on active dynamic IR thermal imaging--a preliminary study.

Proper diagnostic assessment of burn wound depth is of the highest importance in selecting the mode of burn wound treatment. Several diagnostic methods--clinical and histopathological evaluation, as well as methods employing IR imaging--static thermography and active dynamic thermography (ADT)--are compared on the basis of in vivo experiments conducted on three domestic pigs (23 burn wounds). ADT is presented here as a new, reliable and quantitative method of assessing burn wound depth on the basis of discrimination of the thermal properties of burnt tissue. In the case of ADT registration of thermal images was performed following thermal pulse excitation. A series of captured infrared images was used as the basis for calculating the thermal time constant tau for each pixel. The parameter values were compared with histopathological and clinical assessments of burn depth. The mean value of tau was found for burns, which heal within 3 weeks (tau=12.08+/-1.94s) and for burns, which did not heal during this period (tau=9.07+/-0.68s), p<0.05. The accuracy, sensitivity and specificity of all the methods tested were compared, the best results coming from ADT. The ADT method is fast, non-invasive and relatively inexpensive, although it still requires further animal experimentation as well as clinical study to confirm the results.

Telemedicine--education and practice.

Telemedicine is most commonly associated with a "long-distance" surgery in macro or micro dimensions. An example is a doctor operating on a patient who is in space or on a submarine, while he himself is in a telemedicine center found thousands of kilometers away. The basis for telemedicine is the technological progress that we are observing in the last decade. Therefore, there is a need for cooperation between doctors and engineers in the fields of research, education and in offering medical services. Interdisciplinary character of telemedicine requires cooperation especially between medical and technical universities. The article presents already completed and just started telemedicine projects in Poland as a result of cooperation between Medical University of Gdansk and Gdansk University of Technology.