Neonatal Face and Facial Landmark Detection from Video Recordings.

This paper explores automated face and facial landmark detection of neonates, which is an important first step in many video-based neonatal health applications, such as vital sign estimation, pain assessment, sleep-wake classification, and jaundice detection. Utilising three publicly available datasets of neonates in the clinical environment, 366 images (258 subjects) and 89 (66 subjects) were annotated for training and testing, respectively. Transfer learning was applied to two YOLO-based models, with input training images augmented with random horizontal flipping, photo-metric colour distortion, translation and scaling during each training epoch. Additionally, the re-orientation of input images and fusion of trained deep learning models was explored. Our proposed model based on YOLOv7Face outperformed existing methods with a mean average precision of 84.8% for face detection, and a normalised mean error of 0.072 for facial landmark detection. Overall, this will assist in the development of fully automated neonatal health assessment algorithms.Clinical relevance- Accurate face and facial landmark detection provides an automated and non-contact option to assist in video-based neonatal health applications.

Comparison of simultaneous auscultation and ultrasound for clinical assessment of bowel peristalsis in neonates.

Introduction: Assessment of bowel health in ill preterm infants is essential to prevent and diagnose early potentially life-threatening intestinal conditions such as necrotizing enterocolitis. Auscultation of bowel sounds helps assess peristalsis and is an essential component of this assessment. Aim: We aim to compare conventional bowel sound auscultation using acoustic recordings from an electronic stethoscope to real-time bowel motility visualized on point-of-care bowel ultrasound (US) in neonates with no known bowel disease. Methods: This is a prospective observational cohort study in neonates on full enteral feeds with no known bowel disease. A 3M™ Littmann® Model 3200 electronic stethoscope was used to obtain a continuous 60-s recording of bowel sounds at a set region over the abdomen, with a concurrent recording of US using a 12l high-frequency Linear probe. The bowel sounds heard by the first investigator using the stethoscope were contemporaneously transferred for a computerized assessment of their electronic waveforms. The second investigator, blinded to the auscultation findings, obtained bowel US images using a 12l Linear US probe. All recordings were analyzed for bowel peristalsis (duration in seconds) by each of the two methods. Results: We recruited 30 neonates (gestational age range 27-43 weeks) on full enteral feeds with no known bowel disease. The detection of bowel peristalsis (duration in seconds) by both methods (acoustic and US) was reported as a percentage of the total recording time for each participant. Comparing the time segments of bowel sound detection by digital stethoscope recording to that of the visual detection of bowel movements in US revealed a median time of peristalsis with US of 58%, compared to 88.3% with acoustic assessment (p < 0.002). The median regression difference was 26.7% [95% confidence interval (CI) 5%-48%], demonstrating no correlation between the two methods. Conclusion: Our study demonstrates disconcordance between the detection of bowel sounds by auscultation and the detection of bowel motility in real time using US in neonates on full enteral feeds and with no known bowel disease. Better innovative methods using artificial intelligence to characterize bowel sounds, integrating acoustic mapping with sonographic detection of bowel peristalsis, will allow us to develop continuous neonatal bowel sound monitoring devices.

Land use and land cover (LULC) performance modeling using machine learning algorithms: a case study of the city of Melbourne, Australia.

Accurate spatial information on Land use and land cover (LULC) plays a crucial role in city planning. A widely used method of obtaining accurate LULC maps is a classification of the categories, which is one of the challenging problems. Attempts have been made considering spectral (Sp), statistical (St), and index-based (Ind) features in developing LULC maps for city planning. However, no work has been reported to automate LULC performance modeling for their robustness with machine learning (ML) algorithms. In this paper, we design seven schemes and automate the LULC performance modeling with six ML algorithms-Random Forest, Support Vector Machine with Linear kernel, Support Vector Machine with Radial basis function kernel, Artificial Neural Network, Naïve Bayes, and Generalised Linear Model for the city of Melbourne, Australia on Sentinel-2A images. Experimental results show that the Random Forest outperforms remaining ML algorithms in the classification accuracy (0.99) on all schemes. The robustness and statistical analysis of the ML algorithms (for example, Random Forest imparts over 0.99 F1-score for all five categories and p value [Formula: see text] 0.05 from Wilcoxon ranked test over accuracy measures) against varying training splits demonstrate the effectiveness of the proposed schemes. Thus, providing a robust measure of LULC maps in city planning.

Artificial intelligence-driven wearable technologies for neonatal cardiorespiratory monitoring: Part 1 wearable technology.

With the development of Artificial Intelligence techniques, smart health monitoring is becoming more popular. In this study, we investigate the trend of wearable sensors being adopted and developed in neonatal cardiorespiratory monitoring. We performed a search of papers published from the year 2000 onwards. We then reviewed the advances in sensor technologies and wearable modalities for this application. Common wearable modalities included clothing (39%); chest/abdominal belts (25%); and adhesive patches (15%). Popular singular physiological information from sensors included electrocardiogram (15%), breathing (24%), oxygen saturation and photoplethysmography (13%). Many studies (46%) incorporated a combination of these signals. There has been extensive research in neonatal cardiorespiratory monitoring using both single and multi-parameter systems. Poor data quality is a common issue and further research into combining multi-sensor information to alleviate this should be investigated. IMPACT STATEMENT: State-of-the-art review of sensor technology for wearable neonatal cardiorespiratory monitoring. Review of the designs for wearable neonatal cardiorespiratory monitoring. The use of multi-sensor information to improve physiological data quality has been limited in past research. Several sensor technologies have been implemented and tested on adults that have yet to be explored in the newborn population.

Ensemble Approach on Deep and Handcrafted Features for Neonatal Bowel Sound Detection.

For the care of neonatal infants, abdominal auscultation is considered a safe, convenient, and inexpensive method to monitor bowel conditions. With the help of early automated detection of bowel dysfunction, neonatologists could create a diagnosis plan for early intervention. In this article, a novel technique is proposed for automated peristalsis sound detection from neonatal abdominal sound recordings and compared to various other machine learning approaches. It adopts an ensemble approach that utilises handcrafted as well as one and two dimensional deep features obtained from Mel Frequency Cepstral Coefficients (MFCCs). The results are then refined with the help of a hierarchical Hidden Semi-Markov Models (HSMM) strategy. We evaluate our method on abdominal sounds collected from 49 newborn infants admitted to our tertiary Neonatal Intensive Care Unit (NICU). The results of leave-one-patient-out cross validation show that our method provides an accuracy of 95.1% and an Area Under Curve (AUC) of 85.6%, outperforming both the baselines and the recent works significantly. These encouraging results show that our proposed Ensemble-based Deep Learning model is helpful for neonatologists to facilitate tele-health applications.

Noisy Neonatal Chest Sound Separation for High-Quality Heart and Lung Sounds.

Stethoscope-recorded chest sounds provide the opportunity for remote cardio-respiratory health monitoring of neonates. However, reliable monitoring requires high-quality heart and lung sounds. This paper presents novel artificial intelligence-based Non-negative Matrix Factorisation (NMF) and Non-negative Matrix Co-Factorisation (NMCF) methods for neonatal chest sound separation. To assess these methods and compare them with existing single-channel separation methods, an artificial mixture dataset was generated comprising heart, lung, and noise sounds. Signal-to-noise ratios were then calculated for these artificial mixtures. These methods were also tested on real-world noisy neonatal chest sounds and assessed based on vital sign estimation error, and a signal quality score of 1-5, developed in our previous works. Overall, both the proposed NMF and NMCF methods outperform the next best existing method by 2.7 dB to 11.6 dB for the artificial dataset, and 0.40 to 1.12 signal quality improvement for the real-world dataset. The median processing time for the sound separation of a 10 s recording was found to be 28.3 s for NMCF and 342 ms for NMF. With the stable and robust performance of our proposed methods, we believe these methods are useful to denoise neonatal heart and lung sounds in the real-world environment.

Artificial intelligence-driven wearable technologies for neonatal cardiorespiratory monitoring. Part 2: artificial intelligence.

BACKGROUND: With the development of Artificial Intelligence (AI) techniques, smart health monitoring, particularly neonatal cardiorespiratory monitoring with wearable devices, is becoming more popular. To this end, it is crucial to investigate the trend of AI and wearable sensors being developed in this domain. METHODS: We performed a review of papers published in IEEE Xplore, Scopus, and PubMed from the year 2000 onwards, to understand the use of AI for neonatal cardiorespiratory monitoring with wearable technologies. We reviewed the advances in AI development for this application and potential future directions. For this review, we assimilated machine learning (ML) algorithms developed for neonatal cardiorespiratory monitoring, designed a taxonomy, and categorised the methods based on their learning capabilities and performance. RESULTS: For AI related to wearable technologies for neonatal cardio-respiratory monitoring, 63% of studies utilised traditional ML techniques and 35% utilised deep learning techniques, including 6% that applied transfer learning on pre-trained models. CONCLUSIONS: A detailed review of AI methods for neonatal cardiorespiratory wearable sensors is presented along with their advantages and disadvantages. Hierarchical models and suggestions for future developments are highlighted to translate these AI technologies into patient benefit. IMPACT: State-of-the-art review in artificial intelligence used for wearable neonatal cardiorespiratory monitoring. Taxonomy design for artificial intelligence methods. Comparative study of AI methods based on their advantages and disadvantages.

Monkeypox Virus Detection Using Pre-trained Deep Learning-based Approaches.

Monkeypox virus is emerging slowly with the decline of COVID-19 virus infections around the world. People are afraid of it, thinking that it would appear as a pandemic like COVID-19. As such, it is crucial to detect them earlier before widespread community transmission. AI-based detection could help identify them at the early stage. In this paper, we aim to compare 13 different pre-trained deep learning (DL) models for the Monkeypox virus detection. For this, we initially fine-tune them with the addition of universal custom layers for all of them and analyse the results using four well-established measures: Precision, Recall, F1-score, and Accuracy. After the identification of the best-performing DL models, we ensemble them to improve the overall performance using a majority voting over the probabilistic outputs obtained from them. We perform our experiments on a publicly available dataset, which results in average Precision, Recall, F1-score, and Accuracy of 85.44%, 85.47%, 85.40%, and 87.13%, respectively with the help of our proposed ensemble approach. These encouraging results, which outperform the state-of-the-art methods, suggest that the proposed approach is applicable to health practitioners for mass screening.

Prediction of Neonatal Respiratory Distress in Term Babies at Birth from Digital Stethoscope Recorded Chest Sounds.

Neonatal respiratory distress is a common condition that if left untreated, can lead to short- and long-term complications. This paper investigates the usage of digital stethoscope recorded chest sounds taken within 1 min post-delivery, to enable early detection and prediction of neonatal respiratory distress. Fifty-one term newborns were included in this study, 9 of whom developed respiratory distress. For each newborn, 1 min anterior and posterior recordings were taken. These recordings were pre-processed to remove noisy segments and obtain high-quality heart and lung sounds. The random undersampling boosting (RUSBoost) classifier was then trained on a variety of features, such as power and vital sign features extracted from the heart and lung sounds. The RUSBoost algorithm produced specificity, sensitivity, and accuracy results of 85.0%, 66.7% and 81.8%, respectively. Clinical relevance--- This paper investigates the feasibility of digital stethoscope recorded chest sounds for early detection of respiratory distress in term newborn babies, to enable timely treatment and management.

Fruit classification using attention-based MobileNetV2 for industrial applications.

Recent deep learning methods for fruits classification resulted in promising performance. However, these methods are with heavy-weight architectures in nature, and hence require a higher storage and expensive training operations due to feeding a large number of training parameters. There is a necessity to explore lightweight deep learning models without compromising the classification accuracy. In this paper, we propose a lightweight deep learning model using the pre-trained MobileNetV2 model and attention module. First, the convolution features are extracted to capture the high-level object-based information. Second, an attention module is used to capture the interesting semantic information. The convolution and attention modules are then combined together to fuse both the high-level object-based information and the interesting semantic information, which is followed by the fully connected layers and the softmax layer. Evaluation of our proposed method, which leverages transfer learning approach, on three public fruit-related benchmark datasets shows that our proposed method outperforms the four latest deep learning methods with a smaller number of trainable parameters and a superior classification accuracy. Our model has a great potential to be adopted by industries closely related to the fruit growing and retailing or processing chain for automatic fruit identification and classifications in the future.

Fusion of multi-scale bag of deep visual words features of chest X-ray images to detect COVID-19 infection.

Chest X-ray (CXR) images have been one of the important diagnosis tools used in the COVID-19 disease diagnosis. Deep learning (DL)-based methods have been used heavily to analyze these images. Compared to other DL-based methods, the bag of deep visual words-based method (BoDVW) proposed recently is shown to be a prominent representation of CXR images for their better discriminability. However, single-scale BoDVW features are insufficient to capture the detailed semantic information of the infected regions in the lungs as the resolution of such images varies in real application. In this paper, we propose a new multi-scale bag of deep visual words (MBoDVW) features, which exploits three different scales of the 4th pooling layer's output feature map achieved from VGG-16 model. For MBoDVW-based features, we perform the Convolution with Max pooling operation over the 4th pooling layer using three different kernels: [Formula: see text], [Formula: see text], and [Formula: see text]. We evaluate our proposed features with the Support Vector Machine (SVM) classification algorithm on four CXR public datasets (CD1, CD2, CD3, and CD4) with over 5000 CXR images. Experimental results show that our method produces stable and prominent classification accuracy (84.37%, 88.88%, 90.29%, and 83.65% on CD1, CD2, CD3, and CD4, respectively).

Attention-based VGG-16 model for COVID-19 chest X-ray image classification.

Computer-aided diagnosis (CAD) methods such as Chest X-rays (CXR)-based method is one of the cheapest alternative options to diagnose the early stage of COVID-19 disease compared to other alternatives such as Polymerase Chain Reaction (PCR), Computed Tomography (CT) scan, and so on. To this end, there have been few works proposed to diagnose COVID-19 by using CXR-based methods. However, they have limited performance as they ignore the spatial relationships between the region of interests (ROIs) in CXR images, which could identify the likely regions of COVID-19's effect in the human lungs. In this paper, we propose a novel attention-based deep learning model using the attention module with VGG-16. By using the attention module, we capture the spatial relationship between the ROIs in CXR images. In the meantime, by using an appropriate convolution layer (4th pooling layer) of the VGG-16 model in addition to the attention module, we design a novel deep learning model to perform fine-tuning in the classification process. To evaluate the performance of our method, we conduct extensive experiments by using three COVID-19 CXR image datasets. The experiment and analysis demonstrate the stable and promising performance of our proposed method compared to the state-of-the-art methods. The promising classification performance of our proposed method indicates that it is suitable for CXR image classification in COVID-19 diagnosis.

New bag of deep visual words based features to classify chest x-ray images for COVID-19 diagnosis.

PURPOSE: Because the infection by Severe Acute Respiratory Syndrome Coronavirus 2 (COVID-19) causes the Pneumonia-like effect in the lung, the examination of Chest X-Rays (CXR) can help diagnose the disease. For automatic analysis of images, they are represented in machines by a set of semantic features. Deep Learning (DL) models are widely used to extract features from images. General deep features extracted from intermediate layers may not be appropriate to represent CXR images as they have a few semantic regions. Though the Bag of Visual Words (BoVW)-based features are shown to be more appropriate for different types of images, existing BoVW features may not capture enough information to differentiate COVID-19 infection from other Pneumonia-related infections. METHODS: In this paper, we propose a new BoVW method over deep features, called Bag of Deep Visual Words (BoDVW), by removing the feature map normalization step and adding the deep features normalization step on the raw feature maps. This helps to preserve the semantics of each feature map that may have important clues to differentiate COVID-19 from Pneumonia. RESULTS: We evaluate the effectiveness of our proposed BoDVW features in CXR image classification using Support Vector Machine (SVM) to diagnose COVID-19. Our results on four publicly available COVID-19 CXR image datasets (D1, D2, D3, and D4) reveal that our features produce stable and prominent classification accuracy (82.00% on D1, 87.86% on D2, 87.92% on D3, and 83.22% on D4), particularly differentiating COVID-19 infection from other Pneumonia. CONCLUSION: Our method could be a very useful tool for the quick diagnosis of COVID-19 patients on a large scale.

Vector representation based on a supervised codebook for Nepali documents classification.

Document representation with outlier tokens exacerbates the classification performance due to the uncertain orientation of such tokens. Most existing document representation methods in different languages including Nepali mostly ignore the strategies to filter them out from documents before learning their representations. In this article, we propose a novel document representation method based on a supervised codebook to represent the Nepali documents, where our codebook contains only semantic tokens without outliers. Our codebook is domain-specific as it is based on tokens in a given corpus that have higher similarities with the class labels in the corpus. Our method adopts a simple yet prominent representation method for each word, called probability-based word embedding. To show the efficacy of our method, we evaluate its performance in the document classification task using Support Vector Machine and validate against widely used document representation methods such as Bag of Words, Latent Dirichlet allocation, Long Short-Term Memory, Word2Vec, Bidirectional Encoder Representations from Transformers and so on, using four Nepali text datasets (we denote them shortly as A1, A2, A3 and A4). The experimental results show that our method produces state-of-the-art classification performance (77.46% accuracy on A1, 67.53% accuracy on A2, 80.54% accuracy on A3 and 89.58% accuracy on A4) compared to the widely used existing document representation methods. It yields the best classification accuracy on three datasets (A1, A2 and A3) and a comparable accuracy on the fourth dataset (A4). Furthermore, we introduce the largest Nepali document dataset (A4), called NepaliLinguistic dataset, to the linguistic community.

Fusion of whole and part features for the classification of histopathological image of breast tissue.

PURPOSE: Nowadays Computer-Aided Diagnosis (CAD) models, particularly those based on deep learning, have been widely used to analyze histopathological images in breast cancer diagnosis. However, due to the limited availability of such images, it is always tedious to train deep learning models that require a huge amount of training data. In this paper, we propose a new deep learning-based CAD framework that can work with less amount of training data. METHODS: We use pre-trained models to extract image features that can then be used with any classifier. Our proposed features are extracted by the fusion of two different types of features (foreground and background) at two levels (whole-level and part-level). Foreground and background feature to capture information about different structures and their layout in microscopic images of breast tissues. Similarly, part-level and whole-level features capture are useful in detecting interesting regions scattered in high-resolution histopathological images at local and whole image levels. At each level, we use VGG16 models pre-trained on ImageNet and Places datasets to extract foreground and background features, respectively. All features are extracted from mid-level pooling layers of such models. RESULTS: We show that proposed fused features with a Support Vector Classifier (SVM) produce better classification accuracy than recent methods on BACH dataset and our approach is orders of magnitude faster than the best performing recent method (EMS-Net). CONCLUSION: We believe that our method would be another alternative in the diagnosis of breast cancer because of performance and prediction time.