An Expert Diagnostic System to Automatically Identify Asthma and Chronic Obstructive Pulmonary Disease in Clinical Settings

Integrating Artificial Intelligence in the Diagnosis of COPD Globally: A Way Forward

The advancement of artificial intelligence (AI) capabilities has paved the way for a new frontier in medicine, which has the capability to reduce the burden of COPD globally. AI may reduce health care-associated expenses while potentially increasing diagnostic specificity, improving access to early COPD diagnosis, and monitoring COPD progression and subsequent disease management. We evaluated how AI can be integrated into COPD diagnosing globally and leveraged in resource-constrained settings.AI has been explored in diagnosing and phenotyping COPD through auscultation, pulmonary function testing, and imaging. Clinician collaboration with AI has increased the performance of COPD diagnosing and highlights the important role of clinical decision-making in AI integration. Likewise, AI analysis of computer tomography (CT) imaging in large population-based cohorts has increased diagnostic ability, severity classification, and prediction of outcomes related to COPD. Moreover, a multimodality approach with CT imaging, demographic data, and spirometry has been shown to improve machine learning predictions of the progression to COPD compared to each modality alone. Prior research has primarily been conducted in high-income country settings, which may lack generalization to a global population. AI is a World Health Organization priority with the potential to reduce health care barriers in low- and middle-income countries. We recommend a collaboration between clinicians and an AI-supported multimodal approach to COPD diagnosis as a step towards achieving this goal. We believe the interplay of CT imaging, spirometry, biomarkers, and sputum analysis may provide unique insights across settings that could provide a basis for clinical decision-making that includes early intervention for those diagnosed with COPD.

Application of artificial intelligence for the classification of the clinical outcome and therapy in patients with viral infections: The case of COVID-19

BACKGROUND

With the end of the coronavirus disease 2019 (COVID-19) pandemic, it becomes intriguing to observe the impact of innovative digital technologies on the diagnosis and management of diseases, in order to improve clinical outcomes for patients.

OBJECTIVE

The research aims to enhance diagnostics, prediction, and personalized treatment for patients across three classes of clinical severity (mild, moderate, and severe). What sets this study apart is its innovative approach, wherein classification extends beyond mere disease presence, encompassing the classification of disease severity. This novel perspective lays the foundation for a crucial decision support system during patient triage.

METHODS

An artificial neural network, as a deep learning technique, enabled the development of a complex model based on the analysis of data collected during the process of diagnosing and treating 1000 patients at the Tešanj General Hospital, Bosnia and Herzegovina.

RESULTS

The final model achieved a classification accuracy of 82.4% on the validation data set, which testifies to the successful application of the artificial neural network in the classification of clinical outcomes and therapy in patients infected with viral infections.

CONCLUSION

The results obtained show that expert systems are valuable tools for decision support in healthcare in communities with limited resources and increased demands. The research has the potential to improve patient care for future epidemics and pandemics.

A novel multimodal framework for early diagnosis and classification of COPD based on CT scan images and multivariate pulmonary respiratory diseases

BACKGROUND AND OBJECTIVE

Chronic Obstructive Pulmonary Disease (COPD) is one of the world's worst diseases; its early diagnosis using existing methods like statistical machine learning techniques, medical diagnostic tools, conventional medical procedures, and other methods is challenging due to misclassification results of COPD diagnosis and takes a long time to perform accurate prediction. Due to the severe consequences of COPD, detection and accurate diagnosis of COPD at an early stage is essential. This paper aims to design and develop a multimodal framework for early diagnosis and accurate prediction of COPD patients based on prepared Computerized Tomography (CT) scan images and lung sound/cough (audio) samples using machine learning techniques, which are presented in this study.

METHOD

The proposed multimodal framework extracts texture, histogram intensity, chroma, Mel-Frequency Cepstral Coefficients (MFCCs), and Gaussian scale space from the prepared CT images and lung sound/cough samples. Accurate data from All India Institute Medical Sciences (AIIMS), Raipur, India, and the open respiratory CT images and lung sound/cough (audio) sample dataset validate the proposed framework. The discriminatory features are selected from the extracted feature sets using unsupervised ML techniques, and customized ensemble learning techniques are applied to perform early classification and assess the severity levels of COPD patients.

RESULTS

The proposed framework provided 97.50%, 98%, and 95.30% accuracy for early diagnosis of COPD patients based on the fusion technique, CT diagnostic model, and cough sample model.

CONCLUSION

Finally, we compare the performance of the proposed framework with existing methods, current approaches, and conventional benchmark techniques for early diagnosis.

Diagnostic Performance of a Machine Learning Algorithm (Asthma/Chronic Obstructive Pulmonary Disease [COPD] Differentiation Classification) Tool Versus Primary Care Physicians and Pulmonologists in Asthma, COPD, and Asthma/COPD Overlap

BACKGROUND

The differential diagnosis of asthma and chronic obstructive pulmonary disease (COPD) poses a challenge in clinical practice and its misdiagnosis results in inappropriate treatment, increased exacerbations, and potentially death.

OBJECTIVE

To investigate the diagnostic accuracy of the Asthma/COPD Differentiation Classification (AC/DC) tool compared with primary care physicians and pulmonologists in asthma, COPD, and asthma-COPD overlap.

METHODS

The AC/DC machine learning-based diagnostic tool was developed using 12 parameters from electronic health records of more than 400,000 patients aged 35 years and older. An expert panel of three pulmonologists and four general practitioners from five countries evaluated 119 patient cases from a prospective observational study and provided a confirmed diagnosis (n = 116) of asthma (n = 53), COPD (n = 43), asthma-COPD overlap (n = 7), or other (n = 13). Cases were then reviewed by 180 primary care physicians and 180 pulmonologists from nine countries and by the AC/DC tool, and diagnostic accuracies were compared with reference to the expert panel diagnoses.

RESULTS

Average diagnostic accuracy of the AC/DC tool was superior to that of primary care physicians (median difference, 24%; 95% posterior credible interval: 17% to 29%; P < .0001) and was noninferior and superior (median difference, 12%; 95% posterior credible interval: 6% to 17%; P < .0001 for noninferiority and P = .0006 for superiority) to that of pulmonologists. Average diagnostic accuracies were 73%, 50%, and 61% by AC/DC tool, primary care physicians, and pulmonologists versus expert panel diagnosis, respectively.

CONCLUSION

The AC/DC tool demonstrated superior diagnostic accuracy compared with primary care physicians and pulmonologists in the diagnosis of asthma and COPD in patients aged 35 years and greater and has the potential to support physicians in the diagnosis of these conditions in clinical practice.

Classification of Parkinson's disease from smartphone recording data using time-frequency analysis and convolutional neural network

BACKGROUND

Parkinson's disease (PD) is a long-term neurodegenerative disease of the central nervous system. The current diagnosis is dependent on clinical observation and the abilities and experience of a trained specialist. One of the symptoms that affects most patients is voice impairment.

OBJECTIVE

Voice samples are non-invasive data that can be collected remotely for diagnosis and disease progression monitoring. In this study, we analyzed voice recording data from a smartphone as a possible medical self-diagnosis tool by using only one-second voice recording. The data from one of the largest mobile PD studies, the mPower study, was used.

METHODS

A total of 29,798 ten-second voice recordings on smartphone from 4,051 participants were used for the analysis. The voice recordings were from sustained phonation by participants saying /aa/ for ten seconds into an iPhone microphone. A dataset comprising 385,143 short one-second audio samples was generated from the original ten-second voice recordings. The samples were converted to a spectrogram using a short-time Fourier transform. CNN models were then applied to classify the samples.

RESULTS

Classification accuracies of the proposed method with LeNet-5, ResNet-50, and VGGNet-16 are 97.7 ± 0.1%, 98.6 ± 0.2%, and 99.3 ± 0.1%, respectively.

CONCLUSIONS

We achieve a respectable classification performance using a generalized approach on a dataset with a large number of samples. The result emphasizes that an analysis based on one-second clip recorded on a smartphone could be a promising non-invasive and remotely available PD biomarker.

A novel convolution bi-directional gated recurrent unit neural network for emotion recognition in multichannel electroencephalogram signals

BACKGROUND

Recognising emotions in humans is a great challenge in the present era and has several applications under affective computing. Deep learning (DL) found a success tool for predict for emotions in different modalities.

OBJECTIVE

To predict 3D emotions with high accuracy in multichannel physiological signals, i.e. electroencephalogram (EEG).

METHODS

A hybrid DL model consist of CNN and GRU is proposed in this work for emotion recognition in EEG recordings. A convolution neural network (CNN) has the capability of learning abstract representation, whereas gated recurrent units (GRU) have the capability of exploring temporal correlation. A bi-directional variation of GRU is used here to learn features in both directions. Discrete and dimensional emotion indices are recognised in two publicly available datasets namely SEED and DREAMER, respectively. A fused feature of energy and Shannon entropy (𝐸𝑛𝑆𝐸→) and energy and differential entropy (𝐸𝑛𝐷𝐸→) features are fed to the proposed classifier to improve the efficiency of the model.

RESULTS

The performance of the presented model is measured in terms of average accuracy, which is obtained as 86.9% and 93.9% for SEED and DREAMER datasets, respectively.

CONCLUSION

The proposed convolution bi-directional gated recurrent unit neural network (CNN-BiGRU) model outperforms most of the state-of-the-art and competitive hybrid DL models, which indicates the effectiveness of emotion recognition using EEG signals and provides a scientific base for the implementation of human-computer interaction (HCI).

Liver fat analysis using optimized support vector machine with support vector regression

BACKGROUND

Fatty liver disease is a common condition caused by excess fat in the liver. It consists of two types: Alcoholic Fatty Liver Disease, also called alcoholic steatohepatitis, and Non-Alcoholic Fatty Liver Disease (NAFLD). As per epidemiological studies, fatty liver encompasses 9% to 32% of the general population in India and affects overweight people.

OBJECTIVE

An Optimized Support Vector Machine with Support Vector Regression model is proposed to evaluate the volume of liver fat by image analysis (LFA-OSVM-SVR).

METHOD

The input computed tomography (CT) liver images are collected from the Chennai liver foundation and Liver Segmentation (LiTS) datasets. Here, input datasets are pre-processed using Gaussian smoothing filter and bypass filter to reduce noise and improve image intensity. The proposed U-Net method is used to perform the liver segmentation. The Optimized Support Vector Machine is used to classify the liver images as fatty liver image and normal images. The support vector regression (SVR) is utilized for analyzing the fat in percentage.

RESULTS

The LFA-OSVM-SVR model effectively analyzed the liver fat from CT scan images. The proposed approach is activated in python and its efficiency is analyzed under certain performance metrics.

CONCLUSION

The proposed LFA-OSVM-SVR method attains 33.4%, 28.3%, 25.7% improved accuracy with 55%, 47.7%, 32.6% lower error rate for fatty image classification and 30%, 21%, 19.5% improved accuracy with 57.9%, 46.5%, 31.76% lower error rate for normal image classificationthan compared to existing methods such as Convolutional Neural Network (CNN) with Fractional Differential Enhancement (FDE) (CNN-FDE), Fully Convolutional Networks (FCN) and Non-negative Matrix Factorization (NMF) (FCN-NMF), and Deep Learning with Fully Convolutional Networks (FCN) (DL-FCN).

AcneTyper: An automatic diagnosis method of dermoscopic acne image via self-ensemble and stacking

BACKGROUND

Acne is a skin lesion type widely existing in adolescents, and poses computational challenges for automatic diagnosis. Computer vision algorithms are utilized to detect and determine different subtypes of acne. Most of the existing acne detection algorithms are based on the facial natural images, which carry noisy factors like illuminations.

OBJECTIVE

In order to tackle this issue, this study collected a dataset ACNEDer of dermoscopic acne images with annotations. Deep learning methods have demonstrated powerful capabilities in automatic acne diagnosis, and they usually release the training epoch with the best performance as the delivered model.

METHODS

This study proposes a novel self-ensemble and stacking-based framework AcneTyper for diagnosing the acne subtypes. Instead of delivering the best epoch, AcneTyper consolidates the prediction results of all training epochs as the latent features and stacks the best subset of these latent features for distinguishing different acne subtypes.

RESULTS

The proposed AcneTyper framework achieves a promising detection performance of acne subtypes and even outperforms a clinical dermatologist with two-year experiences by 6.8% in accuracy.

CONCLUSION

The method we proposed is used to determine different subtypes of acne and outperforms inexperienced dermatologists and contributes to reducing the probability of misdiagnosis.

Diagnostic value of multi-parameter MRI and colour B-ultrasound elastography in benign and malignant thyroid nodules

BACKGROUND

The early diagnosis of thyroid cancer depends on the popularisation and development of diagnostic imaging techniques and the continuous improvement of physician diagnosis.

OBJECTIVE

To investigate the clinical value of multi-parameter magnetic resonance imaging (MRI) and colour B-ultrasound elastography in thyroid nodules.

METHODS

The clinical and imaging data of 252 patients with thyroid nodules who were admitted to our hospital were collected. All patients underwent preoperative colour B-ultrasound elastography and MRI. The postoperative pathological results were the gold standard for diagnosing benign and malignant thyroid nodules. The accuracy, sensitivity and specificity of MRI, colour B-ultrasound elastography and their combination for diagnosing benign and malignant thyroid nodules were compared.

RESULTS

This study included 252 patients with 388 nodules. There were 169 patients with solitary nodules and 83 patients with multiple nodules. The maximum diameter of the thyroid nodules was 0.32-1.00 (0.75 ± 0.20) cm. The accuracy of MRI diagnosis (348/388) was 89.69%, the sensitivity was 92.98%, and the specificity was 65.22%. The diagnostic accuracy, sensitivity and specificity of colour B-ultrasound elastography (332/388) were 85.57%, 88.30% and 65.22%, respectively. The accuracy rate of combined diagnosis (376/388) was 96.91%, the sensitivity was 98.25%, and the specificity was 86.96%, which was significantly higher than MRI and colour B-ultrasound elastography alone. The area under the curve (AUC) of MRI, colour B-ultrasound elastography and combined diagnosis were 0.768, 0.791 and 0.926, respectively. The AUC of the three diagnostic methods was > 0.7, indicating that the three diagnostic methods had good diagnostic value. The AUC for combined diagnosis was significantly higher than that of MRI and colour B-mode ultrasound elastography alone.

CONCLUSION

Combined ultrasound and MRI have high diagnostic accuracy and specificity for benign and malignant thyroid nodules. This diagnostic method can be applied in clinical practice.

Assessment of artificial intelligence-aided reading in the detection of nasal bone fractures

BACKGROUND: Artificial intelligence (AI) technology is a promising diagnostic adjunct in fracture detection. However, few studies describe the improvement of clinicians’ diagnostic accuracy for nasal bone fractures with the aid of AI technology. OBJECTIVE: This study aims to determine the value of the AI model in improving the diagnostic accuracy for nasal bone fractures compared with manual reading. METHODS: A total of 252 consecutive patients who had undergone facial computed tomography (CT) between January 2020 and January 2021 were enrolled in this study. The presence or absence of a nasal bone fracture was determined by two experienced radiologists. An AI algorithm based on the deep-learning algorithm was engineered, trained and validated to detect fractures on CT images. Twenty readers with various experience were invited to read CT images with or without AI. The accuracy, sensitivity and specificity with the aid of the AI model were calculated by the readers. RESULTS: The deep-learning AI model had 84.78% sensitivity, 86.67% specificity, 0.857 area under the curve (AUC) and a 0.714 Youden index in identifying nasal bone fractures. For all readers, regardless of experience, AI-aided reading had higher sensitivity ([94.00 ± 3.17]% vs [83.52 ± 10.16]%, P< 0.001), specificity ([89.75 ± 6.15]% vs [77.55 ± 11.38]%, P< 0.001) and AUC (0.92 ± 0.04 vs 0.81 ± 0.10, P< 0.001) compared with reading without AI. With the aid of AI, the sensitivity, specificity and AUC were significantly improved in readers with 1–5 years or 6–10 years of experience (all P< 0.05, Table 4). For readers with 11–15 years of experience, no evidence suggested that AI could improve sensitivity and AUC (P= 0.124 and 0.152, respectively). CONCLUSION: The AI model might aid less experienced physicians and radiologists in improving their diagnostic performance for the localisation of nasal bone fractures on CT images.

Exploring the relationship between brain white matter change and higher degree of invisible hand tremor with computer technology

BACKGROUND: At present, the clinical diagnosis of white matter change (WMC) patients depends on cranial magnetic resonance imaging (MRI) technology. This diagnostic method is costly and does not allow for large-scale screening, leading to delays in the patient’s condition due to inability to receive timely diagnosis. OBJECTIVE: To evaluate whether the burden of WMC is associated with the degree of invisible hand tremor in humans. METHODS: Previous studies have shown that tremor is associated with WMC, however, tremor does not always have imaging of WMC. Therefore, to confirm that the appearance of WMC causes tremor, which are sometimes invisible to the naked eye, we achieved an optical-based computer-aided diagnostic device by detecting the invisible hand tremor, and we proposed a calculation method of WMC volume by using the characteristics of MRI images. RESULTS: Statistical analysis results further clarified the relationship between WMC and tremor, and our devices are validated for the detection of tremors with WMC. CONCLUSIONS: The burden of WMC volume is positive factor for degree of invisible hand tremor in the participants without visible hand tremor. Detection technology provides a more convenient and low-cost evaluating method before MRI for tremor diseases.

Association of COVID-19 patient’s condition with fasting blood glucose and body mass index: A retrospective study

BACKGROUND: The COVID-19 pandemic broke out in 2019 and rapidly spread across the globe. Most of the severe and dead cases are middle-aged and elderly patients with chronic systemic diseases. OBJECTIVE: This study aimed to assess the association between fasting blood glucose (FPG) and body mass index (BMI) levels in patients with coronavirus disease 2019 (COVID-19) under different conditions. METHODS: Experimental-related information (age, gender, BMI, and FPG on the second day of admission) from 86 COVID-19 cases (47 males and 39 females) with an average age of (39 ± 17) years was collected in April and November 2020. These cases were divided into three groups according to the most severe classification of each case determined by the clinical early warning indicators of severe-critically illness, the degree of progression, and the treatment plan shown in the diagnosis and treatment plan of COVID-19 pneumonia. Statistical models were used to analyze the differences in the levels of FPG and BMI, age, and gender among the three groups. RESULTS: 1. Experimental group: 21 patients with asymptomatic or and mild symptoms (group A), 45 patients with common non-progression (group B), and 20 patients with common progression and severe symptoms (group C). 2. The age differences among the three groups were statistically significant and elderly patients had a higher risk of severe disease (t= 4.1404, 3.3933, 9.2123, P= 0.0001, 0.0012, 0.0000). There was a higher proportion of females than males in the normal progression and severe disease cases (χ2= 5.512, P= 0.019). 3. The level of FPG was significantly higher in group C than in group A (t= 3.1655, P= 0.0030) and B (t= 2.0212, P= 0.0475). The number of diabetes or IFG in group C was significantly higher than in group A (χ2= 5.979, P= 0.014) and group B (χ2= 6.088, P= 0.014). 4. BMI was significantly higher in group C than in groups A (t= 3.8839, P= 0.0004) and B (t= 3.8188, P= 0.0003). The number of overweight or obese patients in group C was significantly higher than in groups A (χ2= 8.838, P= 0.003) and B (χ2= 10.794, P= 0.001). 5. Patients’ age, gender, and FPG were independent risk factors for COVID-19 disease progression (β= 0.380, 0.191, 0.186; P= 0.000, 0.034, 0.045). CONCLUSION: The levels of FPG and BMI were significantly increased in the population with common progressive and severe COVID-19. FPG and age are independent risk factors for the progression of COVID-19.

Predicting the Level of Respiratory Support in COVID-19 Patients Using Machine Learning

In this paper, a machine learning-based system for the prediction of the required level of respiratory support in COVID-19 patients is proposed. The level of respiratory support is divided into three classes: class 0 which refers to minimal support, class 1 which refers to non-invasive support, and class 2 which refers to invasive support. A two-stage classification system is built. First, the classification between class 0 and others is performed. Then, the classification between class 1 and class 2 is performed. The system is built using a dataset collected retrospectively from 3491 patients admitted to tertiary care hospitals at the University of Louisville Medical Center. The use of the feature selection method based on analysis of variance is demonstrated in the paper. Furthermore, a dimensionality reduction method called principal component analysis is used. XGBoost classifier achieves the best classification accuracy (84%) in the first stage. It also achieved optimal performance in the second stage, with a classification accuracy of 83%.

Artificial intelligence to differentiate asthma from COPD in medico-administrative databases

INTRODUCTION

Discriminating asthma from chronic obstructive pulmonary disease (COPD) using medico-administrative databases is challenging but necessary for medico-economic analyses focusing on respiratory diseases. Artificial intelligence (AI) may improve dedicated algorithms.

OBJECTIVES

To assess performance of different AI-based approaches to distinguish asthmatics from COPD patients in medico-administrative databases where the clinical diagnosis is absent. An "Asthma COPD Overlap" category was defined to further test whether AI can detect complexity.

METHODS

This study included 178,962 patients treated by two "R03" treatment prescriptions at least from January 2016 to December 2018 and managed by either a general practitioner and/or a pulmonologist participating in a permanent longitudinal observatory of prescription in ambulatory medicine (LPD). Clinical diagnoses are available in this database and were used as gold standards to develop diagnostic rules. Three types of AI approaches were explored using data restricted to demographics and treatment dispensations: multinomial regression, gradient boosting and recurrent neural networks (RNN). The best performing model (based on metric properties) was then applied to estimate the size of asthma and COPD populations based on a database (LRx) of treatment dispensations between July, 2018 and June, 2019.

RESULTS

The best models were obtained with the boosting approach and RNN, with an overall accuracy of 68%. Performance metrics were better for asthma than COPD. Based on LRx data, the extrapolated numbers of patients treated for asthma and COPD in France were 3.7 and 1.2 million, respectively. Asthma patients were younger than COPD patients (mean, 49.9 vs. 72.1 years); COPD occurred mostly in men (68%) compared to asthma (33%).

CONCLUSION

AI can provide models with acceptable accuracy to distinguish between asthma, ACO and COPD in medico-administrative databases where the clinical diagnosis is absent. Deep learning and machine learning (RNN) had similar performances in this regard.

Assessment of Multi-Layer Perceptron Neural Network for Pulmonary Function Test’s Diagnosis Using ATS and ERS Respiratory Standard Parameters

The aim of the research work is to investigate the operability of the entire 23 pulmonary function parameters, which are stipulated by the American Thoracic Society (ATS) and the European Respiratory Society (ERS), to design a medical decision support system capable of classifying the pulmonary function tests into normal, obstructive, restrictive, or mixed cases. The 23 respiratory parameters specified by the ATS and the ERS guidelines, obtained from the Pulmonary Function Test (PFT) device, were employed as input features to a Multi-Layer Perceptron (MLP) neural network. Thirteen possible MLP Back Propagation (BP) algorithms were assessed. Three different categories of respiratory diseases were evaluated, namely obstructive, restrictive, and mixed conditions. The framework was applied on 201 PFT examinations: 103 normal and 98 abnormal cases. The PFT decision support system’s outcomes were compared with both the clinical truth (physician decision) and the PFT built-in diagnostic software. It yielded 92–99% and 87–92% accuracies on the training and the test sets, respectively. An 88–94% area under the receiver operating characteristic curve (ROC) was recorded on the test set. The system exceeded the performance of the PFT machine by 9%. All 23 ATS\ERS standard PFT parameters can be used as inputs to design a PFT decision support system, yielding a favorable performance compared with the literature and the PFT machine’s diagnosis program.

Prioritization of healthcare systems during pandemics using Cronbach's measure based fuzzy WASPAS approach

Pandemics are well-known as epidemics that spread globally and cause many illnesses and mortality. Because of globalization, the accelerated occurrence and circulation of new microbes, the infection has emerged and the incidence and movement of new microbes have sped up. Using technological devices to minimize the visit durations, specifying days for handling chronic diseases, subsidy for the staff are the alternatives that can help prevent healthcare systems from collapsing during pandemics. The study aims to define the efficient usage of optimization tools during pandemics to prevent healthcare systems from collapsing. In this study, a new integrated framework with fuzzy information is developed, which attempts to prioritize these alternatives for policymakers. First, rating data are assigned respective fuzzy values using the standard singleton grades. Later, criteria weights are determined by extending Cronbach´s measure to fuzzy context. The measure not only understands data consistency comprehensively, but also takes into consideration the attitudinal characteristics of experts. By this approach, a rational weight vector is obtained for decision-making. Further, an improved Weighted Aggregated Sum Product Assessment (WASPAS) algorithm is put forward for ranking alternatives, which is flexibly considering criteria along with personalized ordering and holistic ordering alternatives. The usefulness of the developed framework is tested with the help of a real case study. Rank values of alternatives when unbiased weights are used is given by 0.741, 0.582, 0.640 with ordering asR 1 ≻ R 3 ≻ R 2 . The sensitivity/comparative analysis reveals the impact of the proposed model as useful in selecting the best alternative for the healthcare systems during pandemics.

Differentiating COPD and Asthma using Quantitative CT Imaging and Machine Learning

There are similarities and differences between chronic obstructive pulmonary disease (COPD) and asthma patients in terms of computed tomography (CT) disease-related features. Our objective was to determine the optimal subset of CT imaging features for differentiating COPD and asthma using machine learning.COPD and asthma patients were recruited from Heidelberg University Hospital. CT was acquired and 93 features were extracted (VIDA Diagnostics): percentage of low-attenuating-areas below -950HU (LAA₉₅₀), LAA₉₅₀ hole count, estimated airway-wall-thickness for a 10 mm internal perimeter airway (Pi10), total-airway-count (TAC), as well as inner/outer perimeter/areas and wall thickness for each of five segmental airways, and the average of those five airways. Hybrid feature selection was used to select the optimum number of features, and support vector machine was used to classify COPD and asthma.Ninety-five participants were included (n=48 COPD; n=47 asthma); there were no differences between COPD and asthma for age (p=0.25) or FEV₁ (p=0.31). In a model including all CT features, the accuracy and F1-score was 80% and 81%, respectively. The top features were: LAA_950, LAA₉₅₀ hole count, average outer and inner airway perimeter, outer and inner airway area RB1, and TAC. In the model with only airway features, the accuracy and F1-score were 66% and 68%, respectively. The top features were: inner area RB1, wall thickness RB1, outer area LB1, TAC LB10, average outer/inner perimeter, Pi10, and TAC.In conclusions, COPD and asthma can be differentiated using machine learning with moderate-high accuracy by a subset of only 7 CT features.

Image- versus histogram-based considerations in semantic segmentation of pulmonary hyperpolarized gas images

PURPOSE

To characterize the differences between histogram-based and image-based algorithms for segmentation of hyperpolarized gas lung images.

METHODS

Four previously published histogram-based segmentation algorithms (ie, linear binning, hierarchical k-means, fuzzy spatial c-means, and a Gaussian mixture model with a Markov random field prior) and an image-based convolutional neural network were used to segment 2 simulated data sets derived from a public (n = 29 subjects) and a retrospective collection (n = 51 subjects) of hyperpolarized 129Xe gas lung images transformed by common MRI artifacts (noise and nonlinear intensity distortion). The resulting ventilation-based segmentations were used to assess algorithmic performance and characterize optimization domain differences in terms of measurement bias and precision.

RESULTS

Although facilitating computational processing and providing discriminating clinically relevant measures of interest, histogram-based segmentation methods discard important contextual spatial information and are consequently less robust in terms of measurement precision in the presence of common MRI artifacts relative to the image-based convolutional neural network.

CONCLUSIONS

Direct optimization within the image domain using convolutional neural networks leverages spatial information, which mitigates problematic issues associated with histogram-based approaches and suggests a preferred future research direction. Further, the entire processing and evaluation framework, including the newly reported deep learning functionality, is available as open source through the well-known Advanced Normalization Tools ecosystem.

Identification of pediatric respiratory diseases using a fine-grained diagnosis system

Respiratory diseases, including asthma, bronchitis, pneumonia, and upper respiratory tract infection (RTI), are among the most common diseases in clinics. The similarities among the symptoms of these diseases precludes prompt diagnosis upon the patients' arrival. In pediatrics, the patients' limited ability in expressing their situation makes precise diagnosis even harder. This becomes worse in primary hospitals, where the lack of medical imaging devices and the doctors' limited experience further increase the difficulty of distinguishing among similar diseases. In this paper, a pediatric fine-grained diagnosis-assistant system is proposed to provide prompt and precise diagnosis using solely clinical notes upon admission, which would assist clinicians without changing the diagnostic process. The proposed system consists of two stages: a test result structuralization stage and a disease identification stage. The first stage structuralizes test results by extracting relevant numerical values from clinical notes, and the disease identification stage provides a diagnosis based on text-form clinical notes and the structured data obtained from the first stage. A novel deep learning algorithm was developed for the disease identification stage, where techniques including adaptive feature infusion and multi-modal attentive fusion were introduced to fuse structured and text data together. Clinical notes from over 12000 patients with respiratory diseases were used to train a deep learning model, and clinical notes from a non-overlapping set of about 1800 patients were used to evaluate the performance of the trained model. The average precisions (AP) for pneumonia, RTI, bronchitis and asthma are 0.878, 0.857, 0.714, and 0.825, respectively, achieving a mean AP (mAP) of 0.819. These results demonstrate that our proposed fine-grained diagnosis-assistant system provides precise identification of the diseases.

Deep learning based respiratory sound analysis for detection of chronic obstructive pulmonary disease

In recent times, technologies such as machine learning and deep learning have played a vital role in providing assistive solutions to a medical domain's challenges. They also improve predictive accuracy for early and timely disease detection using medical imaging and audio analysis. Due to the scarcity of trained human resources, medical practitioners are welcoming such technology assistance as it provides a helping hand to them in coping with more patients. Apart from critical health diseases such as cancer and diabetes, the impact of respiratory diseases is also gradually on the rise and is becoming life-threatening for society. The early diagnosis and immediate treatment are crucial in respiratory diseases, and hence the audio of the respiratory sounds is proving very beneficial along with chest X-rays. The presented research work aims to apply Convolutional Neural Network based deep learning methodologies to assist medical experts by providing a detailed and rigorous analysis of the medical respiratory audio data for Chronic Obstructive Pulmonary detection. In the conducted experiments, we have used a Librosa machine learning library features such as MFCC, Mel-Spectrogram, Chroma, Chroma (Constant-Q) and Chroma CENS. The presented system could also interpret the severity of the disease identified, such as mild, moderate, or acute. The investigation results validate the success of the proposed deep learning approach. The system classification accuracy has been enhanced to an ICBHI score of 93%. Furthermore, in the conducted experiments, we have applied K-fold Cross-Validation with ten splits to optimize the performance of the presented deep learning approach.

A Hybrid Model to Classify Patients with Chronic Obstructive Respiratory Diseases

Over the last decades, an increase in the ageing population and age-related diseases has been observed, with the increase in healthcare costs. As so, new solutions to provide more efficient and affordable support to this group of patients are needed. Such solutions should never discard the user and instead should focus on promoting more healthy lifestyles and provide tools for patients’ active participation in the treatment and management of their diseases. In this concern, the Personal Health Empowerment (PHE) project presented in this paper aims to empower patients to monitor and improve their health, using personal data and technology assisted coaching. The work described in this paper focuses on defining an approach for user modelling on patients with chronic obstructive respiratory diseases using a hybrid modelling approach to identify different groups of users. A classification model with 90.4% prediction accuracy was generated combining agglomerative hierarchical clustering and decision tree classification techniques. Furthermore, this model identified 5 clusters which describe characteristics of 5 different types of users according to 7 generated rules. With the modelling approach defined in this study, a personalized coaching solution will be built considering patients with different necessities and capabilities and adapting the support provided, enabling the recognition of early signs of exacerbations and objective self-monitoring and treatment of the disease. The novel factor of this approach resides in the possibility to integrate personalized coaching technologies adapted to each kind of user within a smartphone-based application resulting in a reliable and affordable alternative for patients to manage their disease.

CoughGAN: Generating Synthetic Coughs that Improve Respiratory Disease Classification

Despite the prevalence of respiratory diseases, their diagnosis by clinicians is challenging. Accurately assessing airway sounds requires extensive clinical training and equipment that may not be easily available. Current methods that automate this diagnosis are hindered by their use of features that require pulmonary function tests. We leverage the audio characteristics of coughs to create classifiers that can distinguish common respiratory diseases in adults. Moreover, we build on recent advances in generative adversarial networks to augment our dataset with cleverly engineered synthetic cough samples for each class of major respiratory disease, to balance and increase our dataset size. We experimented on cough samples collected with a smartphone from 45 subjects in a clinic. Our CoughGAN-improved Support Vector Machine and Random Forest models show up to 76% test accuracy and 83% F1 score in classifying subjects' conditions between healthy and three major respiratory diseases. Adding our synthetic coughs improves the performance we can obtain from a relatively small unbalanced healthcare dataset by boosting the accuracy over 30%. Our data augmentation reduces overfitting and discourages the prediction of a single, dominant class. These results highlight the feasibility of automatic, cough-based respiratory disease diagnosis using smartphones or wearables in the wild.

Neural Network-Based Clinical Prediction System for Identifying the Clinical Effects of Saffron (Crocus sativus L) Supplement Therapy on Allergic Asthma: Model Evaluation Study

Background

Asthma is commonly associated with chronic airway inflammation and is the underlying cause of over a million deaths each year. Crocus sativus L, commonly known as saffron, when used in the form of traditional medicines, has demonstrated anti-inflammatory effects which may be beneficial to individuals with asthma.

Objective

The objective of this study was to develop a clinical prediction system using an artificial neural network to detect the effects of C sativus L supplements on patients with allergic asthma.

Methods

A genetic algorithm–modified neural network predictor system was developed to detect the level of effectiveness of C sativus L using features extracted from the clinical, immunologic, hematologic, and demographic information of patients with asthma. The study included data from men (n=40) and women (n=40) individuals with mild or moderate allergic asthma from 18 to 65 years of age. The aim of the model was to estimate and predict the level of effect of C sativus L supplements on each asthma risk factor and to predict the level of alleviation in patients with asthma. A genetic algorithm was used to extract input features for the clinical prediction system to improve its predictive performance. Moreover, an optimization model was developed for the artificial neural network component that classifies the patients with asthma using C sativus L supplement therapy.

Results

The best overall performance of the clinical prediction system was an accuracy greater than 99% for training and testing data. The genetic algorithm–modified neural network predicted the level of effect with high accuracy for anti–heat shock protein (anti-HSP), high sensitivity C-reactive protein (hs-CRP), forced expiratory volume in the first second of expiration (FEV₁), forced vital capacity (FVC), the ratio of FEV₁/FVC, and forced expiratory flow (FEF_25%-75%) for testing data (anti-HSP: 96.5%; hs-CRP: 98.9%; FEV₁: 98.1%; FVC: 97.5%; FEV₁/FVC ratio: 97%; and FEF_25%-75%: 96.7%, respectively).

Conclusions

The clinical prediction system developed in this study was effective in predicting the effect of C sativus L supplements on patients with allergic asthma. This clinical prediction system may help clinicians to identify early on which clinical factors in asthma will improve over the course of treatment and, in doing so, help clinicians to develop effective treatment plans for patients with asthma.

Artificial neural network identifies nonsteroidal anti-inflammatory drugs exacerbated respiratory disease (N-ERD) cohort

Background

To date, there has been no reliable in vitro test to either diagnose or differentiate nonsteroidal anti‐inflammatory drug (NSAID)–exacerbated respiratory disease (N‐ERD). The aim of the present study was to develop and validate an artificial neural network (ANN) for the prediction of N‐ERD in patients with asthma.

Methods

This study used a prospective database of patients with N‐ERD (n = 121) and aspirin‐tolerant (n = 82) who underwent aspirin challenge from May 2014 to May 2018. Eighteen parameters, including clinical characteristics, inflammatory phenotypes based on sputum cells, as well as eicosanoid levels in induced sputum supernatant (ISS) and urine were extracted for the ANN.

Results

The validation sensitivity of ANN was 94.12% (80.32%‐99.28%), specificity was 73.08% (52.21%‐88.43%), and accuracy was 85.00% (77.43%‐92.90%) for the prediction of N‐ERD. The area under the receiver operating curve was 0.83 (0.71‐0.90).

Conclusions

The designed ANN model seems to have powerful prediction capabilities to provide diagnosis of N‐ERD. Although it cannot replace the gold‐standard aspirin challenge test, the implementation of the ANN might provide an added value for identification of patients with N‐ERD. External validation in a large cohort is needed to confirm our results.

Deep Learning on Computerized Analysis of Chronic Obstructive Pulmonary Disease

GOAL

Chronic obstructive pulmonary disease (COPD) is one of the deadliest diseases in the world. Because COPD is an incurable disease and requires considerable time to be diagnosed even by an experienced specialist, it becomes important to provide analysis abnormalities in simple ways. The aim of the study is comparing multiple machine learning algorithms for the early diagnosis of COPD using multi-channel lung sounds.

METHODS

Deep learning is an efficient machine-learning algorithm, which comprises unsupervised training to reduce optimization and supervised training by a feature-based distribution of classification parameters. This study focuses on analyzing multichannel lung sounds using statistical features of frequency modulations that are extracted using the Hilbert-Huang transform.

RESULTS

Deep learning algorithm was used in the classification stage of the proposed model to separate the patients with COPD and healthy subjects. The proposed DL model with the Hilbert-Huang transform based statistical features was successful in achieving high classification performance rates of 93.67%, 91%, and 96.33% for accuracy, sensitivity, and specificity, respectively.

CONCLUSION

The proposed computerized analysis of the multi-channel lung sounds using DL algorithms provides a standardized assessment with high classification performance.

SIGNIFICANCE

Our study is a pioneer study that directly focuses on the lung sounds to separate COPD and non-COPD patients. Analyzing 12-channel lung sounds gives the advantages of assessing the entire lung obstructions.

Personalized medicine for patients with COPD: where are we?

Chronic airflow limitation is the common denominator of patients with chronic obstructive pulmonary disease (COPD). However, it is not possible to predict morbidity and mortality of individual patients based on the degree of lung function impairment, nor does the degree of airflow limitation allow guidance regarding therapies. Over the last decades, understanding of the factors contributing to the heterogeneity of disease trajectories, clinical presentation, and response to existing therapies has greatly advanced. Indeed, diagnostic assessment and treatment algorithms for COPD have become more personalized. In addition to the pulmonary abnormalities and inhaler therapies, extra-pulmonary features and comorbidities have been studied and are considered essential components of comprehensive disease management, including lifestyle interventions. Despite these advances, predicting and/or modifying the course of the disease remains currently impossible, and selection of patients with a beneficial response to specific interventions is unsatisfactory. Consequently, non-response to pharmacologic and non-pharmacologic treatments is common, and many patients have refractory symptoms. Thus, there is an ongoing urgency for a more targeted and holistic management of the disease, incorporating the basic principles of P4 medicine (predictive, preventive, personalized, and participatory). This review describes the current status and unmet needs regarding personalized medicine for patients with COPD. Also, it proposes a systems medicine approach, integrating genetic, environmental, (micro)biological, and clinical factors in experimental and computational models in order to decipher the multilevel complexity of COPD. Ultimately, the acquired insights will enable the development of clinical decision support systems and advance personalized medicine for patients with COPD.

Acoustic Methods for Pulmonary Diagnosis

Recent developments in sensor technology and computational analysis methods enable new strategies to measure and interpret lung acoustic signals that originate internally, such as breathing or vocal sounds, or are externally introduced, such as in chest percussion or airway insonification. A better understanding of these sounds has resulted in a new instrumentation that allows for highly accurate as well as portable options for measurement in the hospital, in the clinic, and even at home. This review outlines the instrumentation for acoustic stimulation and measurement of the lungs. We first review the fundamentals of acoustic lung signals and the pathophysiology of the diseases that these signals are used to detect. Then, we focus on different methods of measuring and creating signals that have been used in recent research for pulmonary disease diagnosis. These new methods, combined with signal processing and modeling techniques, lead to a reduction in noise and allow improved feature extraction and signal classification. We conclude by presenting the results of human subject studies taking advantage of both the instrumentation and signal processing tools to accurately diagnose common lung diseases. This paper emphasizes the active areas of research within modern lung acoustics and encourages the standardization of future work in this field.