1,721,015 research outputs found
Biostatistics of Cardiac Signals: Theory & Applications
Full text linkL’obiettivo della bioingegneria è lo studio dei fenomeni delle scienze della vita. La statistica è un eccellente strumento per la modellazione, l’analisi, la caratterizzazione e l’interpretazione di questi fenomeni. Scopo di questa tesi di dottorato è quello di combinare le principali tecniche statistiche con l'elaborazione dei segnali cardiaci.
L'importanza delle statistiche nella bioingegneria cardiaca può essere compresa attraverso la loro applicazione; quindi, sono state presentate quattro applicazioni reali.
La prima applicazione è l'Adaptive Thresholding Identification Algorithm (AThrIA), nato per identificare le onde P elettrocardiografiche. AThrIA è l'esempio perfetto di quanto la preelaborazione statistica possa essere importante nella pratica clinica cardiaca.
La seconda applicazione è CTG Analyzer, un'interfaccia che estrae automaticamente le caratteristiche cliniche cardiotocografiche. In tal caso, la statistica diventa lo strumento per valutarne la correttezza delle caratteristiche estratte.
La terza applicazione è eCTG, un software per digitalizzare i segnali cardiotocografici. Combinando l’analisi delle distribuzioni e le tecniche di classificazione, eCTG è un importante esempio dell’utilizzo della statistica nell'elaborazione di immagini e segnali.
Infine, la quarta applicazione è la creazione di classificatori per l’elettrocardiografia seriale basati su deep learning. Questi nuovi e innovativi classificatori rappresentano un esempio di come la classificazione statistica supporta la diagnosi clinica.
In conclusione, questa tesi di dottorato sottolinea l'importanza della statistica nella bioingegneria dei segnali cardiaci. Considerando i risultati e il loro significato clinico, la combinazione di bioingegneria cardiaca e statistica è uno strumento valido per supportare la ricerca scientifica. Legati allo stesso scopo, tali scienze sono in grado di caratterizzare i fenomeni delle scienze della vita, diventando una scienza unica, la biostatistica.Aim of bioengineering is to investigate phenomena of life sciences. Considering that statistic is an excellent tool for modeling, analyzing, characterizing and interpreting phenomena, aim of this doctoral thesis is to merge the major biostatistical techniques and the bioengineering processing of cardiac signals.
The importance of statistics in cardiac bioengineering can be deeply understand through its application; thus, four real applications were presented.
The first is the Adaptive Thresholding Identification Algorithm (AThrIA), born to identify/characterize electrocardiographic P waves. AThrIA is the perfect example of how much statistical preprocessing can be important in cardiac clinical practice.
The second application is CTG Analyzer, an interface that automatically extracts cardiotocographic clinical features. About CTG Analyzer feature extraction, biostatistics is a fundamental instrument to evaluate its correctness.
The third application is eCTG, a software to digitalize cardiotocographic signals from images, using a statistical pixel clustering procedure. Combining distributions analysis and classification, eCTG is an important example of statistics in image/signal processing.
Finally, the fourth application is the creation of deep-learning serial ECG classifiers, specific neural networks to detect cardiac emerging pathology. Based on serial electrocardiography, these new and innovative classifiers represent samples of the real importance of classification in supporting clinical diagnosis.
In conclusion, this doctoral thesis underlines the importance of statistic in bioengineering of cardiac signals. Considering the results and their clinical meaning, the combination of cardiac bioengineering and statistics is a valid instrument to support the scientific research. Linked by the same aim, they are able to quantitative/qualitative characterize the phenomena of life sciences, becoming a single science, biostatistics
Computerized otoscopy image-based artificial intelligence model utilizing deep features provided by vision transformer, grid search optimization, and support vector machine for otitis media diagnosis
No full textOtitis media (OM) is an inflammation of the middle ear, often associated with fluid accumulation and characterized by symptoms such as ear pain, fever, and impaired hearing. Timely and accurate diagnosis of OM is essential to facilitate prompt treatment and mitigate the risk of complications such as hearing loss or chronic infection, particularly in regions with limited access to healthcare professionals. In this study, we introduce an advanced computational model for automated OM diagnosis, utilizing the vision transformer (ViT) architecture to extract highly discriminative features from otoscope images. The proposed approach employs a grid search optimization algorithm in combination with a support vector machine (SVM) classifier to accurately recognize different types of OM based on deep feature representations. All experiments were conducted using a publicly accessible Ear Imagery dataset containing 880 otoscope images, categorized into four distinct classes. As a result, the proposed model demonstrated remarkable efficacy, achieving an impressive accuracy rate of 99.37%. It successfully classified all OM types. At its core, the emergence of advanced computational models in healthcare represents a transformative leap that promises to close gaps in access to medical expertise and revolutionize diagnostic practices. Harnessing the power of machine learning and leveraging vast datasets, these models offer unprecedented accuracy and efficiency, paving the way for early intervention and improving patient outcomes on a global scale
Postural data from Stargardt's syndrome patients
Full text linkThe database is a collection of postural data acquired from 10 patients affected by the rare Stargardt’s syndrome, all having the ABCA4 gene mutation, and from 10 control healthy subjects. Specifically, the database includes a file (.xlxs) called SubjectsData and 20 datasets (MATLAB structures) containing postural signals. Each subject performed a total of 15 postural tests, 5 postural tests for 3 different conditions (‘C’: eyes-closed; ‘O’: eyes-open, still target fixation; ‘M’: eyes-open, moving target tracking). For each postural test, 11 postural derived signals (the anterior-posterior force, the medio-lateral force, the vertical force, the plate moment about x axis, the plate moment about y axis, the plate moment about z axis, the plate moment about top plate surface about x axis, the plate moment about top plat surface about y axis, the x-coordinate of the center of pressure, the y-coordinate of the center of pressure, and the free moment about z axis) were computed from 8 raw signals, acquired at the Ophthalmic Hospital of Turin, Italy, through an 8-channel Kistler 9286A force platform connected to a Step32 system. Thus, a total of 285 postural signals (120 raw and 165 derived) are available for each subject. The database may be useful to: (1) investigate postural adaptations of patients affected by Stargardt’s syndrome; (2) support definition of rehabilitative procedures to reduce postural instability of patients affected by Stargardt’s syndrome; and (3) support investigation on visual control of balance in the general population
Normalization of Electrocardiogram-Derived Cardiac Risk Indices: A Scoping Review of the Open-Access Literature
Full text linkChanges in cardiac function and morphology are reflected in variations in the electrocardiogram (ECG) and, in turn, in the cardiac risk indices derived from it. These variations have led to the introduction of normalization as a step to compensate for possible biasing factors responsible for inter- and intra-subject differences, which can affect the accuracy of ECG-derived risk indices in assessing cardiac risk. The aim of this work is to perform a scoping review to provide a comprehensive collection of open-access published research that examines normalized ECG-derived parameters used as markers of cardiac anomalies or instabilities. The literature search was conducted from February to July 2024 in the major global electronic bibliographic repositories. Overall, 39 studies were selected. Results suggest extensive use of normalization on heart rate variability-related indices (49% of included studies), QT-related indices (18% of included studies), and T-wave alternans (5% of included studies), underscoring their recognized importance and suggesting that normalization may enhance their role as clinically useful risk markers. However, the primary objective of the included studies was not to evaluate the effect of normalization itself; thus, further research is needed to definitively assess the impact and advantages of normalization across various ECG-derived parameters
Wearable and Portable Devices for Acquisition of Cardiac Signals while Practicing Sport: A Scoping Review
Full text linkWearable and portable devices capable of acquiring cardiac signals are at the frontier of the sport industry. They are becoming increasingly popular for monitoring physiological parameters while practicing sport, given the advances in miniaturized technologies, powerful data, and signal processing applications. Data and signals acquired by these devices are increasingly used to monitor athletes’ performances and thus to define risk indices for sport-related cardiac diseases, such as sudden cardiac death. This scoping review investigated commercial wearable and portable devices employed for cardiac signal monitoring during sport activity. A systematic search of the literature was conducted on PubMed, Scopus, and Web of Science. After study selection, a total of 35 studies were included in the review. The studies were categorized based on the application of wearable or portable devices in (1) validation studies, (2) clinical studies, and (3) development studies. The analysis revealed that standardized protocols for validating these technologies are necessary. Indeed, results obtained from the validation studies turned out to be heterogeneous and scarcely comparable, since the metrological characteristics reported were different. Moreover, the validation of several devices was carried out during different sport activities. Finally, results from clinical studies highlighted that wearable devices are crucial to improve athletes’ performance and to prevent adverse cardiovascular events
Multiclass Convolutional Neural Networks for Atrial Fibrillation Classification
No full textAtrial fibrillation (AF) is a common supraventricular arrhythmia. Its automatic identification by standard 12-lead electrocardiography (ECG) is still challenging. Recently, deep learning provided new instruments able to mimic the diagnostic ability of clinicians but only in case of binary classification (AF vs. normal sinus rhythm-NSR). However, binary classification is far from the real scenarios, where AF has to be discriminated also from several other physiological and pathological conditions. The aim of this work is to present a new AF multiclass classifier based on a convolutional neural network (CNN), able to discriminate AF from NSR, premature atrial contraction (PAC) and premature ventricular contraction (PVC). Overall, 2796 12-lead ECG recordings were selected from the open-source "PhysioNet/Computing in Cardiology Challenge 2021" database, to construct a dataset constituted by four balanced classes, namely AF class, PAC class, PVC class, and NSR class. Each lead of each ECG recording was decomposed into spectrogram by continuous wavelet transform and saved as 2D grayscale images, used to feed a 6-layers CNN. Considering the same CNN architecture, a multiclass classifiers (all classes) and three binary classifiers (AF class, PAC class, and PVC class vs. NSR class) were created and validated by a stratified shuffle split cross-validation of 10 splits. Performance was quantified in terms of area under the curve (AUC) of the receiver operating characteristic. Multiclass classifier performance was high (AF class: 96.6%; PAC class: 95.3%; PVC class: 92.8%; NSR class: 97.4%) and preferable to binary classifiers. Thus, our CNN AF multiclass classifier proved to be an efficient tool for AF discrimination from physiological and pathological confounders. Clinical Relevance-Our CNN AF multiclass classifier proved to be suitable for AF discrimination in real scenarios
Digital cardiotocography: What is the optimal sampling frequency?
No full textCardiotocography (CTG) is the most popular prenatal diagnostic test for establishing fetal health and consists in simultaneous recording of fetal heart rate (FHR, bpm) and maternal uterine contraction (UC, mmHg) traces. Typically, FHR and UC traces are visually analyzed and interpreted by clinicians. Recently, software applications like CTG Analyzer have been developed to support visual CTG interpretation by making it more objective and independent from clinician’s experience. Automatic CTG analysis requires CTG-traces digitalization and thus assessment of a correct sampling frequency (SF). Thus, this paper aims to investigate dependency of automatic CTG analysis on SF in order to identify optimal SF (OSF) for FHR and UC traces that minimizes computational efforts without jeopardizing CTG interpretation. To this aim, the “CTU-CHB intra-partum CTG database” was considered and visually annotated by an expert gynecologist. FHR and UC traces, originally sampled at 4 Hz, were down sampled at 2 Hz, 1 Hz, 0.4 Hz and 0.2 Hz, and automatically analyzed using CTG Analyzer. Eventually, results obtained through automatic analysis were compared to visual annotations, which were taken as reference. A cumulative statistical index (CSI), ranging from 0.00% to 100.00%, was defined as a linear combination of positive-predictive value, sensitivity, false-positive rate and false-negative rate. OSF was defined as the one that maximizes CSI. If CSI was showing the same value for more than one SF, the lowest SF was selected as the optimal since minimizing computational efforts. Results indicate that OSF for FHR is 2 Hz (CSI ≥ 85.41%), whereas OSF for UC is 0.2 Hz (CSI = 75.21%)
The role of central vision in posture: Postural sway adaptations in Stargardt patients
Full text linkThe role of central and peripheral vision in the maintenance of upright stance is debated in literature. Stargardt disease causes visual deficits affecting the central field, but leaving unaltered a patient's peripheral vision. Hence, the study of this rare pathology gives the opportunity to selectively investigate the role of central vision in posture. Postural sway in quiet stance was analyzed in 10 Stargardt patients and 10 control subjects, in three different conditions: (1) eyes closed, (2) eyes open, gazing at a fixed target, and (3) eyes open, tracking a moving target. Stargardt patients outperformed controls in the condition with eyes closed, showing a reduced root mean square (RMS) of the medio-lateral COP displacement, while their performance was not significantly different from controls in the antero- posterior direction. There were no significant differences between patients and controls in open eyes conditions. These results suggest that Stargardt patients adapted to a different visual-somatosensory integration, relying less on vision, especially in the medio-lateral direction. Hence, the central vision seems to affect mostly the medio-lateral direction of postural sway. This finding supports the plausibility of the ‘‘functional sensitivity hypothesis'', that assigns complementary roles to central and peripheral vision in the control of posture
The Prognostic Value of Electrocardiographic Alternans in the Primary Prevention on Patients Having an Implantable Cardioverter Defibrillator
No full textPrimary prevention therapy with implantable cardioverter defibrillator (ICD) may benefit in specificity evaluating the left ventricular ejection fraction together with other electrocardiogram (ECG)-derived cardiac risk indexes, such as ECG alternans (ECGA). ECGA, the ABAB morphology fluctuation of ECG waves (P wave/QRS complex/T wave), results in P-wave/QRS/T-wave alternans (PWA/QRSA/TWA). This work aims to validate ECGA prognostic value on ECGs acquired from ICD patients (Leiden University Medical Center Database). Thus, 82 controls (ICD therapy was not needed) and 40 cases (ICD therapy was needed) were enrolled. ECGA was analyzed by the enhanced adaptive matched filter method at rest and exercise. Median ECGA features (amplitude-Am, duration-D, area-Ar, magnitude-M) were computed over leads and ICD groups. ECGA ability to discriminate between ICD groups was evaluated by the area under the curve (AUC) of the receiver operating characteristic curve. Precordial leads allowed a better discrimination than all leads (higher AUC). QRSA, at both rest (controls/cases: Am=5/10 μV, D=33/45 beats, Ar=360/760 μV·ms, M=351/603 μV·beats) and exercise (controls/cases: Am=16/20 μV, D=55/57 beats, Ar=1280/1600 μV·ms, M=992/1316 μV·beats) has the best discriminant power, with AUC values higher than 0.7 at rest. ECGA feature normalization by the ECG mean amplitude was also considered. We can conclude that (1) ECGA predictive power is best expressed at rest, and QRSA seems to be the best ECGA form to identify patients who should benefit from primary prevention ICD therapy, (2) normalization seems not to improve our results
- …
