Bioinformatics. 2024 Jun 17:btae378. doi: 10.1093/bioinformatics/btae378. Online ahead of print.

ABSTRACT

MOTIVATION: ADP-ribosylation is a critical modification involved in regulating diverse cellular processes, including chromatin structure regulation, RNA transcription, and cell death. Bacterial ADP-ribosyltransferase toxins (bARTTs) serve as potent virulence factors that orchestrate the manipulation of host cell functions to facilitate bacterial pathogenesis. Despite their pivotal role, the bioinformatic identification of novel bARTTs poses a formidable challenge due to limited verified data and the inherent sequence diversity among bARTT members.

RESULTS: We proposed a deep learning-based model, ARTNet, specifically engineered to predict bARTTs from bacterial genomes. Initially, we introduced an effective data augmentation method to address the issue of data scarcity in training ARTNet. Subsequently, we employed a data optimization strategy by utilizing ART-related domain subsequences instead of the primary full sequences, thereby significantly enhancing the performance of ARTNet. ARTNet achieved a Matthew's correlation coefficient (MCC) of 0.9351 and an F1-score (macro) of 0.9666 on repeated independent test datasets, outperforming three other deep learning models and six traditional machine learning models in terms of time efficiency and accuracy. Furthermore, we empirically demonstrated the ability of ARTNet to predict novel bARTTs across domain superfamilies without sequence similarity. We anticipate that ARTNet will greatly facilitate the screening and identification of novel bARTTs from bacterial genomes.

AVAILABILITY: ARTNet is publicly accessible at http://www.mgc.ac.cn/ARTNet/. The source code of ARTNet is freely available at https://github.com/zhengdd0422/ARTNet/.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:38885365 | DOI:10.1093/bioinformatics/btae378

PLoS One. 2024 Jun 17;19(6):e0292026. doi: 10.1371/journal.pone.0292026. eCollection 2024.

ABSTRACT

Machine learning (ML) and deep learning (DL) models are being increasingly employed for medical imagery analyses, with both approaches used to enhance the accuracy of classification/prediction in the diagnoses of various cancers, tumors and bloodborne diseases. To date however, no review of these techniques and their application(s) within the domain of white blood cell (WBC) classification in blood smear images has been undertaken, representing a notable knowledge gap with respect to model selection and comparison. Accordingly, the current study sought to comprehensively identify, explore and contrast ML and DL methods for classifying WBCs. Following development and implementation of a formalized review protocol, a cohort of 136 primary studies published between January 2006 and May 2023 were identified from the global literature, with the most widely used techniques and best-performing WBC classification methods subsequently ascertained. Studies derived from 26 countries, with highest numbers from high-income countries including the United States (n = 32) and The Netherlands (n = 26). While WBC classification was originally rooted in conventional ML, there has been a notable shift toward the use of DL, and particularly convolutional neural networks (CNN), with 54.4% of identified studies (n = 74) including the use of CNNs, and particularly in concurrence with larger datasets and bespoke features e.g., parallel data pre-processing, feature selection, and extraction. While some conventional ML models achieved up to 99% accuracy, accuracy was shown to decrease in concurrence with decreasing dataset size. Deep learning models exhibited improved performance for more extensive datasets and exhibited higher levels of accuracy in concurrence with increasingly large datasets. Availability of appropriate datasets remains a primary challenge, potentially resolvable using data augmentation techniques. Moreover, medical training of computer science researchers is recommended to improve current understanding of leucocyte structure and subsequent selection of appropriate classification models. Likewise, it is critical that future health professionals be made aware of the power, efficacy, precision and applicability of computer science, soft computing and artificial intelligence contributions to medicine, and particularly in areas like medical imaging.

PMID:38885231 | DOI:10.1371/journal.pone.0292026

PLoS One. 2024 Jun 17;19(6):e0304798. doi: 10.1371/journal.pone.0304798. eCollection 2024.

ABSTRACT

Drug-drug interaction (DDI) is the combined effects of multiple drugs taken together, which can either enhance or reduce each other's efficacy. Thus, drug interaction analysis plays an important role in improving treatment effectiveness and patient safety. It has become a new challenge to use computational methods to accelerate drug interaction time and reduce its cost-effectiveness. The existing methods often do not fully explore the relationship between the structural information and the functional information of drug molecules, resulting in low prediction accuracy for drug interactions, poor generalization, and other issues. In this paper, we propose a novel method, which is a deep graph contrastive learning model for drug-drug interaction prediction (DeepGCL for brevity). DeepGCL incorporates a contrastive learning component to enhance the consistency of information between different views (molecular structure and interaction network), which means that the DeepGCL model predicts drug interactions by integrating molecular structure features and interaction network topology features. Experimental results show that DeepGCL achieves better performance than other methods in all datasets. Moreover, we conducted many experiments to analyze the necessity of each component of the model and the robustness of the model, which also showed promising results. The source code of DeepGCL is freely available at https://github.com/jzysj/DeepGCL.

PMID:38885206 | DOI:10.1371/journal.pone.0304798

IEEE Trans Pattern Anal Mach Intell. 2024 Jun 17;PP. doi: 10.1109/TPAMI.2024.3415387. Online ahead of print.

ABSTRACT

Deep learning-based solutions have achieved impressive performance in semantic segmentation but often require large amounts of training data with fine-grained annotations. To alleviate such requisition, a variety of weakly supervised annotation strategies have been proposed, among which scribble supervision is emerging as a popular one due to its user-friendly annotation way. However, the sparsity and diversity of scribble annotations make it nontrivial to train a network to produce deterministic and consistent predictions directly. To address these issues, in this paper we propose holistic solutions involving the design of network structure, loss and training procedure, named CC4S to improve Certainty and Consistency for Scribble-Supervised Semantic Segmentation. Specifically, to reduce uncertainty, CC4S embeds a random walk module into the network structure to make neural representations uniformly distributed within similar semantic regions, which works together with a soft entropy loss function to force the network to produce deterministic predictions. To encourage consistency, CC4S adopts self-supervision training and imposes the consistency loss on the eigenspace of the probability transition matrix in the random walk module (we named neural eigenspace). Such self-supervision inherits the category-level discriminability from the neural eigenspace and meanwhile helps the network focus on producing consistent predictions for the salient parts and neglect semantically heterogeneous backgrounds. Finally, to further improve the performance, CC4S uses the network predictions as pseudo-labels and retrains the network with an extra color constraint regularizer on pseudo-labels to boost semantic consistency in color space. Rich experiments demonstrate the excellent performance of CC4S. In particular, under scribble supervision, CC4S achieves comparable performance to those from fully supervised methods. Comprehensive ablation experiments verify the effectiveness of the design choices in CC4S and its robustness under extreme supervision cases, i.e., when scribbles are shrunk proportionally or dropped randomly. The code for this work has been open-sourced at https://github.com/panzhiyi/CC4S.

PMID:38885110 | DOI:10.1109/TPAMI.2024.3415387

IEEE J Biomed Health Inform. 2024 Jun 17;PP. doi: 10.1109/JBHI.2024.3415479. Online ahead of print.

ABSTRACT

Cough is an important symptom in children with acute and chronic respiratory disease. Daily cough is common in Cystic Fibrosis (CF) and increased cough is a symptom of pulmonary exacerbation. To date, cough assessment is primarily subjective in clinical practice and research. Attempts to develop objective, automatic cough counting tools have faced reliability issues in noisy environments and practical barriers limiting long-term use. This single-center pilot study evaluated usability, acceptability and performance of a mechanoacoustic sensor (MAS), previously used for cough classification in adults, in 36 children with CF over brief and multi-day periods in four cohorts. Children whose health was at baseline and who had symptoms of pulmonary exacerbation were included. We trained, validated, and deployed custom deep learning algorithms for accurate cough detection and classification from other vocalization or artifacts with an overall area under the receiver-operator characteristic curve (AUROC) of 0.96 and average precision (AP) of 0.93. Child and parent feedback led to a redesign of the MAS towards a smaller, more discreet device acceptable for daily use in children. Additional improvements optimized power efficiency and data management. The MAS's ability to objectively measure cough and other physiologic signals across clinic, hospital, and home settings is demonstrated, particularly aided by an AUROC of 0.97 and AP of 0.96 for motion artifact rejection. Examples of cough frequency and physiologic parameter correlations with participant-reported outcomes and clinical measurements for individual patients are presented. The MAS is a promising tool in objective longitudinal evaluation of cough in children with CF.

PMID:38885105 | DOI:10.1109/JBHI.2024.3415479

IEEE J Biomed Health Inform. 2024 Jun 17;PP. doi: 10.1109/JBHI.2024.3415000. Online ahead of print.

ABSTRACT

Learning-based methods offer performance leaps over traditional methods in classification analysis of high-dimensional functional MRI (fMRI) data. In this domain, deep-learning models that analyze functional connectivity (FC) features among brain regions have been particularly promising. However, many existing models receive as input temporally static FC features that summarize inter-regional interactions across an entire scan, reducing the temporal sensitivity of classifiers by limiting their ability to leverage information on dynamic FC features of brain activity. To improve the performance of baseline classification models without compromising efficiency, here we propose a novel plug-in based on a graph neural network, GraphCorr, to provide enhanced input features to baseline models. The proposed plug-in computes a set of latent FC features with enhanced temporal information while maintaining comparable dimensionality to static features. Taking brain regions as nodes and blood-oxygen-level-dependent (BOLD) signals as node inputs, GraphCorr leverages a node embedder module based on a transformer encoder to capture dynamic latent representations of BOLD signals. GraphCorr also leverages a lag filter module to account for delayed interactions across nodes by learning correlational features of windowed BOLD signals across time delays. These two feature groups are then fused via a message passing algorithm executed on the formulated graph. Comprehensive demonstrations on three public datasets indicate improved classification performance for several state-of-the-art graph and convolutional baseline models when they are augmented with GraphCorr.

PMID:38885104 | DOI:10.1109/JBHI.2024.3415000

IEEE Trans Biomed Circuits Syst. 2024 Jun 17;PP. doi: 10.1109/TBCAS.2024.3415392. Online ahead of print.

ABSTRACT

Surface electromyography (sEMG) is a State-of-the-Art (SoA) sensing modality for non-invasive human-machine interfaces for consumer, industrial, and rehabilitation use cases. The main limitation of the current sEMG-driven control policies is the sEMG's inherent variability, especially cross-session due to sensor repositioning; this limits the generalization of the Machine/Deep Learning (ML/DL) in charge of the signal-to-command mapping. The other hot front on the ML/DL side of sEMG-driven control is the shift from the classification of fixed hand positions to the regression of hand kinematics and dynamics, promising a more versatile and fluid control. We present an incremental online-training strategy for sEMG-based estimation of simultaneous multi-finger forces, using a small Temporal Convolutional Network suitable for embedded learning-on-device. We validate our method on the HYSER dataset, cross-day. Our incremental online training reaches a cross-day Mean Absolute Error (MAE) of (9.58 ± 3.89)% of the Maximum Voluntary Contraction on HYSER's RANDOM dataset of improvised, non-predefined force sequences, which is the most challenging and closest to real scenarios. This MAE is on par with an accuracy-oriented, non-embeddable offline training exploiting more epochs. Further, we demonstrate that our online training approach can be deployed on the GAP9 ultra-low power microcontroller, obtaining a latency of 1.49 ms and an energy draw of just 40.4 uJ per forward-backward-update step. These results show that our solution fits the requirements for accurate and real-time incremental training-on-device.

PMID:38885102 | DOI:10.1109/TBCAS.2024.3415392

IEEE Trans Biomed Circuits Syst. 2024 Jun 17;PP. doi: 10.1109/TBCAS.2024.3415360. Online ahead of print.

ABSTRACT

Electrical capacitance tomography (ECT) can be used to predict information about the interior volume of an object based on measured capacitance at its boundaries. Here, we present a microscale capacitance tomography system with a spatial resolution of 10 microns using an active CMOS microelectrode array. We introduce a deep learning model for reconstructing 3-D volumes of cell cultures using the boundary capacitance measurements acquired from the sensor array, which is trained using a multi-objective loss function that combines a pixel-wise loss function, a distribution-based loss function, and a region-based loss function to improve model's reconstruction accuracy. The multi-objective loss function enhances the model's reconstruction accuracy by 3.2% compared to training only with a pixel-wise loss function. Compared to baseline computational methods, our model achieves an average of 4.6% improvement on the datasets evaluated. We demonstrate our approach on experimental datasets of bacterial biofilms, showcasing the system's ability to resolve microscopic spatial features of cell cultures in three dimensions. Microscale capacitance tomography can be a low-cost, low-power, label-free tool for 3-D imaging of biological samples.

PMID:38885101 | DOI:10.1109/TBCAS.2024.3415360

IEEE Trans Neural Netw Learn Syst. 2024 Jun 17;PP. doi: 10.1109/TNNLS.2024.3409563. Online ahead of print.

ABSTRACT

Multispectral image (MS) and panchromatic image (PAN) fusion, which is also named as multispectral pansharpening, aims to obtain MS with high spatial resolution and high spectral resolution. However, due to the usual neglect of noise and blur generated in the imaging and transmission phases of data during training, many deep learning (DL) pansharpening methods fail to perform on the dataset containing noise and blur. To tackle this problem, a variational optimization-guided two-stage network (VOGTNet) for multispectral pansharpening is proposed in this work, and the performance of variational optimization (VO)-based pansharpening methods relies on prior information and estimates of spatial-spectral degradation from the target image to other two original images. Concretely, we propose a dual-branch fusion network (DBFN) based on supervised learning and train it by using the datasets containing noise and blur to generate the prior fusion result as the prior information that can remove noise and blur in the initial stage. Subsequently, we exploit the estimated spectral response function (SRF) and point spread function (PSF) to simulate the process of spatial-spectral degradation, respectively, thereby making the prior fusion result and the adaptive recovery model (ARM) jointly perform unsupervised learning on the original dataset to restore more image details and results in the generation of the high-resolution MSs in the second stage. Experimental results indicate that the proposed VOGTNet improves pansharpening performance and shows strong robustness against noise and blur. Furthermore, the proposed VOGTNet can be extended to be a general pansharpening framework, which can improve the ability to resist noise and blur of other supervised learning-based pansharpening methods. The source code is available at https://github.com/HZC-1998/VOGTNet.

PMID:38885100 | DOI:10.1109/TNNLS.2024.3409563

Med Biol Eng Comput. 2024 Jun 17. doi: 10.1007/s11517-024-03148-2. Online ahead of print.

ABSTRACT

Parkinson's disease (PD) is a degenerative nervous system disorder involving motor disturbances. Motor alterations affect the gait according to the progression of PD and can be used by experts in movement disorders to rate the severity of the disease. However, this rating depends on the expertise of the clinical specialist. Therefore, the diagnosis may be inaccurate, particularly in the early stages of PD where abnormal gait patterns can result from normal aging or other medical conditions. Consequently, several classification systems have been developed to enhance PD diagnosis. In this paper, a PD gait severity classification algorithm was developed using vertical ground reaction force (VGRF) signals. The VGRF records used are from a public database that includes 93 PD patients and 72 healthy controls adults. The work presented here focuses on modeling each foot's gait stance phase signals using a modified convolutional long deep neural network (CLDNN) architecture. Subsequently, the results of each model are combined to predict PD severity. The classifier performance was evaluated using ten-fold cross-validation. The best-weighted accuracies obtained were 99.296(0.128)% and 99.343(0.182)%, with the Hoehn-Yahr and UPDRS scales, respectively, outperforming previous results presented in the literature. The classifier proposed here can effectively differentiate gait patterns of different PD severity levels based on gait signals of the stance phase.

PMID:38884852 | DOI:10.1007/s11517-024-03148-2

Vis Comput Ind Biomed Art. 2024 Jun 17;7(1):15. doi: 10.1186/s42492-024-00166-7.

ABSTRACT

Skin lesion classification plays a crucial role in the early detection and diagnosis of various skin conditions. Recent advances in computer-aided diagnostic techniques have been instrumental in timely intervention, thereby improving patient outcomes, particularly in rural communities lacking specialized expertise. Despite the widespread adoption of convolutional neural networks (CNNs) in skin disease detection, their effectiveness has been hindered by the limited size and data imbalance of publicly accessible skin lesion datasets. In this context, a two-step hierarchical binary classification approach is proposed utilizing hybrid machine and deep learning (DL) techniques. Experiments conducted on the International Skin Imaging Collaboration (ISIC 2017) dataset demonstrate the effectiveness of the hierarchical approach in handling large class imbalances. Specifically, employing DenseNet121 (DNET) as a feature extractor and random forest (RF) as a classifier yielded the most promising results, achieving a balanced multiclass accuracy (BMA) of 91.07% compared to the pure deep-learning model (end-to-end DNET) with a BMA of 88.66%. The RF ensemble exhibited significantly greater efficiency than other machine-learning classifiers in aiding DL to address the challenge of learning with limited data. Furthermore, the implemented predictive hybrid hierarchical model demonstrated enhanced performance while significantly reducing computational time, indicating its potential efficiency in real-world applications for the classification of skin lesions.

PMID:38884841 | DOI:10.1186/s42492-024-00166-7

Ann Biomed Eng. 2024 Jun 17. doi: 10.1007/s10439-024-03559-0. Online ahead of print.

ABSTRACT

Machine learning (ML) has led to significant advances in dentistry, easing the workload of professionals and improving the performance of various medical processes. The fields of periodontology and implantology can profit from these advances for tasks such as determining periodontally compromised teeth, assisting doctors in the implant planning process, determining types of implants, or predicting the occurrence of peri-implantitis. The current paper provides an overview of recent ML techniques applied in periodontology and implantology, aiming to identify popular models for different medical tasks, to assess the impact of the training data on the success of the automatic algorithms and to highlight advantages and disadvantages of various approaches. 48 original research papers, published between 2016 and 2023, were selected and divided into four classes: periodontology, implant planning, implant brands and types, and success of dental implants. These papers were analyzed in terms of aim, technical details, characteristics of training and testing data, results, and medical observations. The purpose of this paper is not to provide an exhaustive survey, but to show representative methods from recent literature that highlight the advantages and disadvantages of various approaches, as well as the potential of applying machine learning in dentistry.

PMID:38884831 | DOI:10.1007/s10439-024-03559-0

Phys Eng Sci Med. 2024 Jun 17. doi: 10.1007/s13246-024-01441-w. Online ahead of print.

ABSTRACT

An automated scoring system for cleanliness assessment during video capsule endoscopy (VCE) is presently lacking. The present study focused on developing an approach to automatically assess the cleanliness in VCE frames as per the latest scoring i.e., Korea-Canada (KODA). Initially, an easy-to-use mobile application called artificial intelligence-KODA (AI-KODA) score was developed to collect a multi-label image dataset of twenty-eight patient capsule videos. Three readers (gastroenterology fellows), who had been trained in reading VCE, rated this dataset in a duplicate manner. The labels were saved automatically in real-time. Inter-rater and intra-rater reliability were checked. The developed dataset was then randomly split into train:validate:test ratio of 70:20:10 and 60:20:20. It was followed by a comprehensive benchmarking and evaluation of three multi-label classification tasks using ten machine learning and two deep learning algorithms. Reliability estimation was found to be overall good among the three readers. Overall, random forest classifier achieved the best evaluation metrics, followed by Adaboost, KNeighbours, and Gaussian naive bayes in the machine learning-based classification tasks. Deep learning algorithms outperformed the machine learning-based classification tasks for only VM labels. Thorough analysis indicates that the proposed approach has the potential to save time in cleanliness assessment and is user-friendly for research and clinical use. Further research is required for the improvement of intra-rater reliability of KODA, and the development of automated multi-task classification in this field.

PMID:38884670 | DOI:10.1007/s13246-024-01441-w

Phys Eng Sci Med. 2024 Jun 17. doi: 10.1007/s13246-024-01423-y. Online ahead of print.

ABSTRACT

This study aimed to evaluate the impact of radiation dose and focal spot size on the image quality of super-resolution deep-learning reconstruction (SR-DLR) in comparison with iterative reconstruction (IR) and normal-resolution DLR (NR-DLR) algorithms for cardiac CT. Catphan-700 phantom was scanned on a 320-row scanner at six radiation doses (small and large focal spots at 1.4-4.3 and 5.8-8.8 mGy, respectively). Images were reconstructed using hybrid-IR, model-based-IR, NR-DLR, and SR-DLR algorithms. Noise properties were evaluated through plotting noise power spectrum (NPS). Spatial resolution was quantified with task-based transfer function (TTF); Polystyrene, Delrin, and Bone-50% inserts were used for low-, intermediate, and high-contrast spatial resolution. The detectability index (d') was calculated. Image noise, noise texture, edge sharpness of low- and intermediate-contrast objects, delineation of fine high-contrast objects, and overall quality of four reconstructions were visually ranked. Results indicated that among four reconstructions, SR-DLR yielded the lowest noise magnitude and NPS peak, as well as the highest average NPS frequency, TTF50%, d' values, and visual rank at each radiation dose. For all reconstructions, the intermediate- to high-contrast spatial resolution was maximized at 4.3 mGy, while the lowest noise magnitude and highest d' were attained at 8.8 mGy. SR-DLR at 4.3 mGy exhibited superior noise performance, intermediate- to high-contrast spatial resolution, d' values, and visual rank compared to the other reconstructions at 8.8 mGy. Therefore, SR-DLR may yield superior diagnostic image quality and facilitate radiation dose reduction compared to the other reconstructions, particularly when combined with small focal spot scanning.

PMID:38884668 | DOI:10.1007/s13246-024-01423-y

Rev Sci Instrum. 2024 Jun 1;95(6):063702. doi: 10.1063/5.0206155.

ABSTRACT

Gas cameras are primarily used to detect gas leaks, but their use has been increasingly extended to mass flow quantification (qOGI). We employ the previously published active illuminated gas camera [Bergau et al. "Real-time active-gas imaging of small gas leaks," J. Sens. Sens. Syst. 12, 61-68 (2023) and Bergau et al. "Flow rate quantification of small methane leaks using laser spectroscopy and deep learning," Process Saf. Environ. Prot. 182, 752-759 (2024)] in a real-world application for quantification, enhancing the camera with two new features: sensitivity adaptation and camera-gas distance detection. This technology was applied to a gas leak found in the pressure swing adsorption room of a biogas plant in Germany. We compare its performance with state-of-the-art quantification gas cameras (qOGI), such as Sensia Mileva 33. Such a comparison between active and passive gas cameras is possible for the first time due to the introduced sensitivity tuning. Additionally, we enclosed the gas leak and measure the methane concentration with a flame ionization detector, providing a gold standard for comparison. Our findings revealed relative offsets to our gold standard of -57% and +319% for the DAS-camera and the Sensia, respectively, suggesting that the accuracy of mass flow quantification could be improved through the use of active gas cameras.

PMID:38884562 | DOI:10.1063/5.0206155

Transl Vis Sci Technol. 2024 Jun 3;13(6):10. doi: 10.1167/tvst.13.6.10.

ABSTRACT

PURPOSE: To explore the structural-functional loss relationship from optic-nerve-head- and macula-centred spectral-domain (SD) Optical Coherence Tomography (OCT) images in the full spectrum of glaucoma patients using deep-learning methods.

METHODS: A cohort comprising 5238 unique eyes classified as suspects or diagnosed with glaucoma was considered. All patients underwent ophthalmologic examination consisting of standard automated perimetry (SAP), macular OCT, and peri-papillary OCT on the same day. Deep learning models were trained to estimate G-pattern visual field (VF) mean deviation (MD) and cluster MD using retinal thickness maps from seven layers: retinal nerve fiber layer (RNFL), ganglion cell layer and inner plexiform layer (GCL + IPL), inner nuclear layer and outer plexiform layer (INL + OPL), outer nuclear layer (ONL), photoreceptors and retinal pigmented epithelium (PR + RPE), choriocapillaris and choroidal stroma (CC + CS), total retinal thickness (RT).

RESULTS: The best performance on MD prediction is achieved by RNFL, GCL + IPL and RT layers, with R2 scores of 0.37, 0.33, and 0.31, respectively. Combining macular and peri-papillary scans outperforms single modality prediction, achieving an R2 value of 0.48. Cluster MD predictions show promising results, notably in central clusters, reaching an R2 of 0.56.

CONCLUSIONS: The combination of multiple modalities, such as optic-nerve-head circular B-scans and retinal thickness maps from macular SD-OCT images, improves the performance of MD and cluster MD prediction. Our proposed model demonstrates the highest level of accuracy in predicting MD in the early-to-mid stages of glaucoma.

TRANSLATIONAL RELEVANCE: Objective measures recorded with SD-OCT can optimize the number of visual field tests and improve individualized glaucoma care by adjusting VF testing frequency based on deep-learning estimates of functional damage.

PMID:38884547 | DOI:10.1167/tvst.13.6.10

Acta Orthop. 2024 Jun 17;95:319-324. doi: 10.2340/17453674.2024.40905.

ABSTRACT

BACKGROUND AND PURPOSE: Knowledge concerning the use AI models for the classification of glenohumeral osteoarthritis (GHOA) and avascular necrosis (AVN) of the humeral head is lacking. We aimed to analyze how a deep learning (DL) model trained to identify and grade GHOA on plain radiographs performs. Our secondary aim was to train a DL model to identify and grade AVN on plain radiographs.

PATIENTS AND METHODS: A modified ResNet-type network was trained on a dataset of radiographic shoulder examinations from a large tertiary hospital. A total of 7,139 radiographs were included. The dataset included various projections of the shoulder, and the network was trained using stochastic gradient descent. Performance evaluation metrics, area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to assess the network's performance for each outcome.

RESULTS: The network demonstrated AUC values ranging from 0.73 to 0.93 for GHOA classification and > 0.90 for all AVN classification classes. The network exhibited lower AUC for mild cases compared with definitive cases of GHOA. When none and mild grades were combined, the AUC increased, suggesting difficulties in distinguishing between these 2 grades.

CONCLUSION: We found that a DL model can be trained to identify and grade GHOA on plain radiographs. Furthermore, we show that a DL model can identify and grade AVN on plain radiographs. The network performed well, particularly for definitive cases of GHOA and any level of AVN. However, challenges remain in distinguishing between none and mild GHOA grades.

PMID:38884536 | DOI:10.2340/17453674.2024.40905

J Acoust Soc Am. 2024 Jun 1;155(6):3833-3847. doi: 10.1121/10.0026218.

ABSTRACT

For cochlear implant (CI) listeners, holding a conversation in noisy and reverberant environments is often challenging. Deep-learning algorithms can potentially mitigate these difficulties by enhancing speech in everyday listening environments. This study compared several deep-learning algorithms with access to one, two unilateral, or six bilateral microphones that were trained to recover speech signals by jointly removing noise and reverberation. The noisy-reverberant speech and an ideal noise reduction algorithm served as lower and upper references, respectively. Objective signal metrics were compared with results from two listening tests, including 15 typical hearing listeners with CI simulations and 12 CI listeners. Large and statistically significant improvements in speech reception thresholds of 7.4 and 10.3 dB were found for the multi-microphone algorithms. For the single-microphone algorithm, there was an improvement of 2.3 dB but only for the CI listener group. The objective signal metrics correctly predicted the rank order of results for CI listeners, and there was an overall agreement for most effects and variances between results for CI simulations and CI listeners. These algorithms hold promise to improve speech intelligibility for CI listeners in environments with noise and reverberation and benefit from a boost in performance when using features extracted from multiple microphones.

PMID:38884525 | DOI:10.1121/10.0026218

Eur Rev Med Pharmacol Sci. 2024 Jun;28(11):3702-3710. doi: 10.26355/eurrev_202406_36375.

ABSTRACT

OBJECTIVE: Monitoring Jackson Pratt and Hemovac drains plays a crucial role in assessing a patient's recovery and identifying potential postoperative complications. Accurate and regular monitoring of the blood volume in the drain is essential for making decisions about patient care. However, transferring blood to a measuring cup and recording it is a challenging task for both patients and doctors, exposing them to bloodborne pathogens such as the human immunodeficiency virus (HIV), hepatitis B virus (HBV), and hepatitis C virus (HCV). To automate the recording process with a non-contact approach, we propose an innovative approach that utilizes deep learning techniques to detect a drain in a photograph, compute the blood level in the drain, estimate the blood volume, and display the results on both web and mobile interfaces.

MATERIALS AND METHODS: Our system employs semantic segmentation on images taken with mobile phones to effectively isolate the blood-filled portion of the drain from the rest of the image and compute the blood volume. These results are then sent to mobile and web applications for convenient access. To validate the accuracy and effectiveness of our system, we collected the Drain Dataset, which consists of 1,004 images taken under various background and lighting conditions.

RESULTS: With an average error rate of less than 5% in milliliters, our proposed approach achieves highly accurate blood level detection and estimation, as demonstrated by our trials on this dataset. The system also exhibits robustness to variations in lighting conditions and drain shapes, ensuring its applicability in different clinical scenarios.

CONCLUSIONS: The proposed automated blood volume estimation system can significantly reduce the time and effort required for manual measurements, enabling healthcare professionals to focus on other critical tasks. The dataset and annotations are available at: https://www.kaggle.com/datasets/ayenahin/liquid-volume-detection-from-drain-images and the code for the web application is available at https://github.com/itsjustaplant/AwesomeProject.git.

PMID:38884505 | DOI:10.26355/eurrev_202406_36375

PeerJ. 2024 Jun 11;12:e16538. doi: 10.7717/peerj.16538. eCollection 2024.

ABSTRACT

The cultivation of cashew crops carries numerous economic advantages, and countries worldwide that produce this crop face a high demand. The effects of wind speed and wind direction on crop yield prediction using proficient deep learning algorithms are less emphasized or researched. We propose a combination of advanced deep learning techniques, specifically focusing on long short-term memory (LSTM) and random forest models. We intend to enhance this ensemble model using dynamic time warping (DTW) to assess the spatiotemporal data (wind speed and wind direction) similarities within Jaman North, Jaman South, and Wenchi with their respective production yield. In the Bono region of Ghana, these three areas are crucial for cashew production. The LSTM-DTW-RF model with wind speed and wind direction achieved an R2 score of 0.847 and the LSTM-RF model without these two key features R2 score of (0.74). Both models were evaluated using the augmented Dickey-Fuller (ADF) test, which is commonly used in time series analysis to assess stationarity, where the LSTM-DTW-RF achieved a 90% level of confidence, while LSTM-RF attained an 87.99% level. Among the three municipalities, Jaman South had the highest evaluation scores for the model, with an RMSE of 0.883, an R2 of 0.835, and an MBE of 0.212 when comparing actual and predicted values for Wenchi. In terms of the annual average wind direction, Jaman North recorded (270.5 SW°), Jaman South recorded (274.8 SW°), and Wenchi recorded (272.6 SW°). The DTW similarity distance for the annual average wind speed across these regions fell within specific ranges: Jaman North (±25.72), Jaman South (±25.89), and Wenchi (±26.04). Following the DTW similarity evaluation, Jaman North demonstrated superior performance in wind speed, while Wenchi excelled in wind direction. This underscores the potential efficiency of DTW when incorporated into the analysis of environmental factors affecting crop yields, given its invariant nature. The results obtained can guide further exploration of DTW variations in combination with other machine learning models to predict higher cashew yields. Additionally, these findings emphasize the significance of wind speed and direction in vertical farming, contributing to informed decisions for sustainable agricultural growth and development.

PMID:38881862 | PMC:PMC11177857 | DOI:10.7717/peerj.16538