Ophthalmol Ther. 2024 May 11. doi: 10.1007/s40123-024-00958-3. Online ahead of print.

ABSTRACT

The integration of artificial intelligence (AI) in ophthalmology has promoted the development of the discipline, offering opportunities for enhancing diagnostic accuracy, patient care, and treatment outcomes. This paper aims to provide a foundational understanding of AI applications in ophthalmology, with a focus on interpreting studies related to AI-driven diagnostics. The core of our discussion is to explore various AI methods, including deep learning (DL) frameworks for detecting and quantifying ophthalmic features in imaging data, as well as using transfer learning for effective model training in limited datasets. The paper highlights the importance of high-quality, diverse datasets for training AI models and the need for transparent reporting of methodologies to ensure reproducibility and reliability in AI studies. Furthermore, we address the clinical implications of AI diagnostics, emphasizing the balance between minimizing false negatives to avoid missed diagnoses and reducing false positives to prevent unnecessary interventions. The paper also discusses the ethical considerations and potential biases in AI models, underscoring the importance of continuous monitoring and improvement of AI systems in clinical settings. In conclusion, this paper serves as a primer for ophthalmologists seeking to understand the basics of AI in their field, guiding them through the critical aspects of interpreting AI studies and the practical considerations for integrating AI into clinical practice.

PMID:38734807 | DOI:10.1007/s40123-024-00958-3

Sci Rep. 2024 May 11;14(1):10830. doi: 10.1038/s41598-024-60592-8.

ABSTRACT

De-mining operations are of critical importance for humanitarian efforts and safety in conflict-affected regions. In this paper, we address the challenge of enhancing the accuracy and efficiency of mine detection systems. We present an innovative Deep Learning architecture tailored for pulse induction-based Metallic Mine Detectors (MMD), so called DL-MMD. Our methodology leverages deep neural networks to distinguish amongst nine distinct materials with an exceptional validation accuracy of 93.5%. This high level of precision enables us not only to differentiate between anti-personnel mines, without metal plates but also to detect minuscule 0.2-g vertical paper pins in both mineralized soil and non-mineralized environments. Moreover, through comparative analysis, we demonstrate a substantial 3% and 7% improvement (approx.) in accuracy performance compared to the traditional K-Nearest Neighbors and Support Vector Machine classifiers, respectively. The fusion of deep neural networks with the pulse induction-based MMD not only presents a cost-effective solution but also significantly expedites decision-making processes in de-mining operations, ultimately contributing to improved safety and effectiveness in these critical endeavors.

PMID:38734748 | DOI:10.1038/s41598-024-60592-8

Eye (Lond). 2024 May 11. doi: 10.1038/s41433-024-03085-2. Online ahead of print.

ABSTRACT

BACKGROUND/OBJECTIVES: Artificial intelligence can assist with ocular image analysis for screening and diagnosis, but it is not yet capable of autonomous full-spectrum screening. Hypothetically, false-positive results may have unrealized screening potential arising from signals persisting despite training and/or ambiguous signals such as from biomarker overlap or high comorbidity. The study aimed to explore the potential to detect clinically useful incidental ocular biomarkers by screening fundus photographs of hypertensive adults using diabetic deep learning algorithms.

SUBJECTS/METHODS: Patients referred for treatment-resistant hypertension were imaged at a hospital unit in Perth, Australia, between 2016 and 2022. The same 45° colour fundus photograph selected for each of the 433 participants imaged was processed by three deep learning algorithms. Two expert retinal specialists graded all false-positive results for diabetic retinopathy in non-diabetic participants.

RESULTS: Of the 29 non-diabetic participants misclassified as positive for diabetic retinopathy, 28 (97%) had clinically useful retinal biomarkers. The models designed to screen for fewer diseases captured more incidental disease. All three algorithms showed a positive correlation between severity of hypertensive retinopathy and misclassified diabetic retinopathy.

CONCLUSIONS: The results suggest that diabetic deep learning models may be responsive to hypertensive and other clinically useful retinal biomarkers within an at-risk, hypertensive cohort. Observing that models trained for fewer diseases captured more incidental pathology increases confidence in signalling hypotheses aligned with using self-supervised learning to develop autonomous comprehensive screening. Meanwhile, non-referable and false-positive outputs of other deep learning screening models could be explored for immediate clinical use in other populations.

PMID:38734746 | DOI:10.1038/s41433-024-03085-2

Sci Rep. 2024 May 11;14(1):10801. doi: 10.1038/s41598-024-61561-x.

ABSTRACT

The non-perfusion area (NPA) of the retina is an important indicator in the visual prognosis of patients with branch retinal vein occlusion (BRVO). However, the current evaluation method of NPA, fluorescein angiography (FA), is invasive and burdensome. In this study, we examined the use of deep learning models for detecting NPA in color fundus images, bypassing the need for FA, and we also investigated the utility of synthetic FA generated from color fundus images. The models were evaluated using the Dice score and Monte Carlo dropout uncertainty. We retrospectively collected 403 sets of color fundus and FA images from 319 BRVO patients. We trained three deep learning models on FA, color fundus images, and synthetic FA. As a result, though the FA model achieved the highest score, the other two models also performed comparably. We found no statistical significance in median Dice scores between the models. However, the color fundus model showed significantly higher uncertainty than the other models (p < 0.05). In conclusion, deep learning models can detect NPAs from color fundus images with reasonable accuracy, though with somewhat less prediction stability. Synthetic FA stabilizes the prediction and reduces misleading uncertainty estimates by enhancing image quality.

PMID:38734727 | DOI:10.1038/s41598-024-61561-x

Sci Rep. 2024 May 11;14(1):10812. doi: 10.1038/s41598-024-61063-w.

ABSTRACT

Cervical cancer, the second most prevalent cancer affecting women, arises from abnormal cell growth in the cervix, a crucial anatomical structure within the uterus. The significance of early detection cannot be overstated, prompting the use of various screening methods such as Pap smears, colposcopy, and Human Papillomavirus (HPV) testing to identify potential risks and initiate timely intervention. These screening procedures encompass visual inspections, Pap smears, colposcopies, biopsies, and HPV-DNA testing, each demanding the specialized knowledge and skills of experienced physicians and pathologists due to the inherently subjective nature of cancer diagnosis. In response to the imperative for efficient and intelligent screening, this article introduces a groundbreaking methodology that leverages pre-trained deep neural network models, including Alexnet, Resnet-101, Resnet-152, and InceptionV3, for feature extraction. The fine-tuning of these models is accompanied by the integration of diverse machine learning algorithms, with ResNet152 showcasing exceptional performance, achieving an impressive accuracy rate of 98.08%. It is noteworthy that the SIPaKMeD dataset, publicly accessible and utilized in this study, contributes to the transparency and reproducibility of our findings. The proposed hybrid methodology combines aspects of DL and ML for cervical cancer classification. Most intricate and complicated features from images can be extracted through DL. Further various ML algorithms can be implemented on extracted features. This innovative approach not only holds promise for significantly improving cervical cancer detection but also underscores the transformative potential of intelligent automation within the realm of medical diagnostics, paving the way for more accurate and timely interventions.

PMID:38734714 | DOI:10.1038/s41598-024-61063-w

BMC Med Imaging. 2024 May 11;24(1):107. doi: 10.1186/s12880-024-01292-7.

ABSTRACT

This study addresses the critical challenge of detecting brain tumors using MRI images, a pivotal task in medical diagnostics that demands high accuracy and interpretability. While deep learning has shown remarkable success in medical image analysis, there remains a substantial need for models that are not only accurate but also interpretable to healthcare professionals. The existing methodologies, predominantly deep learning-based, often act as black boxes, providing little insight into their decision-making process. This research introduces an integrated approach using ResNet50, a deep learning model, combined with Gradient-weighted Class Activation Mapping (Grad-CAM) to offer a transparent and explainable framework for brain tumor detection. We employed a dataset of MRI images, enhanced through data augmentation, to train and validate our model. The results demonstrate a significant improvement in model performance, with a testing accuracy of 98.52% and precision-recall metrics exceeding 98%, showcasing the model's effectiveness in distinguishing tumor presence. The application of Grad-CAM provides insightful visual explanations, illustrating the model's focus areas in making predictions. This fusion of high accuracy and explainability holds profound implications for medical diagnostics, offering a pathway towards more reliable and interpretable brain tumor detection tools.

PMID:38734629 | DOI:10.1186/s12880-024-01292-7

Methods. 2024 May 9:S1046-2023(24)00112-9. doi: 10.1016/j.ymeth.2024.05.002. Online ahead of print.

ABSTRACT

Studies have shown that protein glycosylation in cells reflects the real-time dynamics of biological processes, and the occurrence and development of many diseases are closely related to protein glycosylation. Abnormal protein glycosylation can be used as a potential diagnostic and prognostic marker of a disease, as well as a therapeutic target and a new breakthrough point for exploring pathogenesis. To address the issue of significant differences in the prediction results of previous models for different species, we constructed a hybrid deep learning model N-GlycoPred on the basis of dual-layer convolution, a paired attention mechanism and BiLSTM for accurate identification of N-glycosylation sites. By adopting one-hot encoding or the AAindex, we specifically selected the optimum combination of features and deep learning frameworks for human and mouse to refine the models. Based on six independent test datasets, our N-GlycoPred model achieved an average AUC of 0.9553, which is 0.23% higher than MusiteDeep. The comparison results indicate that our model can serve as a powerful tool for N-glycosylation site prescreening for biological researchers.

PMID:38734394 | DOI:10.1016/j.ymeth.2024.05.002

Spectrochim Acta A Mol Biomol Spectrosc. 2024 May 1;317:124396. doi: 10.1016/j.saa.2024.124396. Online ahead of print.

ABSTRACT

Accurate prediction of the concentration of a large number of hyaluronic acid (HA) samples under temperature perturbations can facilitate the rapid determination of HA's appropriate applications. Near-infrared (NIR) spectroscopy analysis combined with deep learning presents an effective solution to this challenge, with current research in this area being scarce. Initially, we introduced a novel feature fusion method based on an intersection strategy and used two-dimensional correlation spectroscopy (2DCOS) and Aquaphotomics to interpret the interaction information in HA solutions reflected by the fused features. Subsequently, we created an innovative, multi-strategy improved Walrus Optimization Algorithm (MIWaOA) for parameter optimization of the deep extreme learning machine (DELM). The final constructed MIWaOA-DELM model demonstrated superior performance compared to partial least squares (PLS), extreme learning machine (ELM), DELM, and WaOA-DELM models. The results of this study can provide a reference for the quantitative analysis of biomacromolecules in complex systems.

PMID:38733911 | DOI:10.1016/j.saa.2024.124396

J Environ Manage. 2024 May 10;360:121097. doi: 10.1016/j.jenvman.2024.121097. Online ahead of print.

ABSTRACT

With high-frequency data of nitrate (NO3-N) concentrations in waters becoming increasingly important for understanding of watershed system behaviors and ecosystem managements, the accurate and economic acquisition of high-frequency NO3-N concentration data has become a key point. This study attempted to use coupled deep learning neural networks and routine monitored data to predict hourly NO3-N concentrations in a river. The hourly NO3-N concentration at the outlet of the Oyster River watershed in New Hampshire, USA, was predicted through neural networks with a hybrid model architecture coupling the Convolutional Neural Networks and the Long Short-Term Memory model (CNN-LSTM). The routine monitored data (the river depth, water temperature, air temperature, precipitation, specific conductivity, pH and dissolved oxygen concentrations) for model training were collected from a nested high-frequency monitoring network, while the high-frequency NO3-N concentration data obtained at the outlet were not included as inputs. The whole dataset was separated into training, validation, and testing processes according to the ratio of 5:3:2, respectively. The hybrid CNN-LSTM model with different input lengths (1d, 3d, 7d, 15d, 30d) displayed comparable even better performance than other studies with lower frequencies, showing mean values of the Nash-Sutcliffe Efficiency 0.60-0.83. Models with shorter input lengths demonstrated both the higher modeling accuracy and stability. The water level, water temperature and pH values at monitoring sites were main controlling factors for forecasting performances. This study provided a new insight of using deep learning networks with a coupled architecture and routine monitored data for high-frequency riverine NO3-N concentration forecasting and suggestions about strategies about variable and input length selection during preprocessing of input data.

PMID:38733844 | DOI:10.1016/j.jenvman.2024.121097

J Environ Manage. 2024 May 10;360:121089. doi: 10.1016/j.jenvman.2024.121089. Online ahead of print.

ABSTRACT

Baseflow is a crucial water source in the inland river basins of high-cold mountainous region, playing a significant role in maintaining runoff stability. It is challenging to select the most suitable baseflow separation method in data-scarce high-cold mountainous region and to evaluate effects of climate factors and underlying surface changes on baseflow variability and seasonal distribution characteristics. Here we attempt to address how meteorological factors and underlying surface changes affect baseflow using the Grey Wolf Optimizer Digital Filter Method (GWO-DFM) for rapid baseflow separation and the Long Short-Term Memory (LSTM) neural network model for baseflow prediction, clarifying interpretability of the LSTM model in baseflow forecasting. The proposed method was successfully implemented using a 63-year time series (1958-2020) of flow data from the Tai Lan River (TLR) basin in the high-cold mountainous region, along with 21 years of ERA5-land meteorological data and MODIS data (2000-2020). The results indicate that: (1) GWO-DFM can rapidly identify the optimal filtering parameters. It employs the arithmetic average of three methods, namely Chapman, Chapman-Maxwell and Eckhardt filter, as the best baseflow separation approach for the TLR basin. Additionally, the baseflow significantly increases after the second mutation of the baseflow rate. (2) Baseflow sources are mainly influenced by precipitation infiltration, glacier frozen soil layers, and seasonal ponding. (3) Solar radiation, temperature, precipitation, and NDVI are the primary factors influencing baseflow changes, with Nash-Sutcliffe efficiency coefficients exceeding 0.78 in both the LSTM model training and prediction periods. (4) Changes in baseflow are most influenced by solar radiation, temperature, and NDVI. This study systematically analyzes the changes in baseflow and response mechanisms in high-cold mountainous region, contributing to the management of water resources in mountainous basins under changing environmental conditions.

PMID:38733842 | DOI:10.1016/j.jenvman.2024.121089

Int J Med Inform. 2024 Apr 26;188:105462. doi: 10.1016/j.ijmedinf.2024.105462. Online ahead of print.

ABSTRACT

OBJECTIVE: For ICD-10 coding causes of death in France in 2018 and 2019, predictions by deep neural networks (DNNs) are employed in addition to fully automatic batch coding by a rule-based expert system and to interactive coding by the coding team focused on certificates with a special public health interest and those for which DNNs have a low confidence index.

METHODS: Supervised seq-to-seq DNNs are trained on previously coded data to ICD-10 code multiple causes and underlying causes of death. The DNNs are then used to target death certificates to be sent to the coding team and to predict multiple causes and underlying causes of death for part of the certificates. Hence, the coding campaign for 2018 and 2019 combines three modes of coding and a loop of interaction between the three.

FINDINGS: In this campaign, 62% of the certificates are automatically batch coded by the expert system, 3% by the coding team, and the remainder by DNNs. Compared to a traditional campaign that would have relied on automatic batch coding and manual coding, the present campaign reaches an accuracy of 93.4% for ICD-10 coding of the underlying cause (95.6% at the European shortlist level). Some limitations (risks of under- or overestimation) appear for certain ICD categories, with the advantage of being quantifiable.

CONCLUSION: The combination of the three coding methods illustrates how artificial intelligence, automated and human codings are mutually enriching. Quantified limitations on some chapters of ICD codes encourage an increase in the volume of certificates sent for manual coding from 2021 onward.

PMID:38733641 | DOI:10.1016/j.ijmedinf.2024.105462

Br J Radiol. 2024 May 11:tqae094. doi: 10.1093/bjr/tqae094. Online ahead of print.

ABSTRACT

OBJECTIVES: This study aimed to assess the impact of super-resolution deep learning reconstruction (SR-DLR) on coronary computed tomography angiography (CCTA) image quality and blooming artifacts from coronary artery stents in comparison to conventional methods, including hybrid iterative reconstruction (HIR) and deep learning-based reconstruction (DLR).

METHODS: A retrospective analysis included sixty-six CCTA patients from July to November 2022. Major coronary arteries were evaluated for image noise, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR). Stent sharpness was quantified using 10-90% edge rise slope (ERS) and 10-90% edge rise distance (ERD). Qualitative analysis employed a 5-point scoring system to assess overall image quality, image noise, vessel wall, and stent structure.

RESULTS: SR-DLR demonstrated significantly lower image noise compared to HIR and DLR. SNR and CNR were notably higher in SR-DLR. Stent ERS was significantly improved in SR-DLR, with mean ERD values of 0.70 ± 0.20 mm for SR-DLR, 1.13 ± 0.28 mm for HIR, and 0.85 ± 0.26 mm for DLR. Qualitatively, SR-DLR scored higher in all categories.

CONCLUSIONS: SR-DLR produces images with lower image noise, leading to improved overall image quality, compared with HIR and DLR. SR-DLR is a valuable image reconstruction algorithm for enhancing the spatial resolution and sharpness of coronary artery stents without being constrained by hardware limitations.

ADVANCED IN KNOWLEDGE: The overall image quality was significantly higher in SR-DLR, resulting in sharper coronary artery stents compared to HIR and DLR.

PMID:38733576 | DOI:10.1093/bjr/tqae094

J Chem Inf Model. 2024 May 11. doi: 10.1021/acs.jcim.3c02088. Online ahead of print.

ABSTRACT

Revealing the mechanisms that influence transcription factor binding specificity is the key to understanding gene regulation. In previous studies, DNA double helix structure and one-hot embedding have been used successfully to design computational methods for predicting transcription factor binding sites (TFBSs). However, DNA sequence as a kind of biological language, the method of word embedding representation in natural language processing, has not been considered properly in TFBS prediction models. In our work, we integrate different types of features of DNA sequence to design a multichanneled deep learning framework, namely MulTFBS, in which independent one-hot encoding, word embedding encoding, which can incorporate contextual information and extract the global features of the sequences, and double helix three-dimensional structural features have been trained in different channels. To extract sequence high-level information effectively, in our deep learning framework, we select the spatial-temporal network by combining convolutional neural networks and bidirectional long short-term memory networks with attention mechanism. Compared with six state-of-the-art methods on 66 universal protein-binding microarray data sets of different transcription factors, MulTFBS performs best on all data sets in the regression tasks, with the average R2 of 0.698 and the average PCC of 0.833, which are 5.4% and 3.2% higher, respectively, than the suboptimal method CRPTS. In addition, we evaluate the classification performance of MulTFBS for distinguishing bound or unbound regions on TF ChIP-seq data. The results show that our framework also performs well in the TFBS classification tasks.

PMID:38733561 | DOI:10.1021/acs.jcim.3c02088

Interdiscip Sci. 2024 May 11. doi: 10.1007/s12539-024-00619-w. Online ahead of print.

ABSTRACT

Accumulating studies have demonstrated close relationships between long non-coding RNAs (lncRNAs) and diseases. Identification of new lncRNA-disease associations (LDAs) enables us to better understand disease mechanisms and further provides promising insights into cancer targeted therapy and anti-cancer drug design. Here, we present an LDA prediction framework called GEnDDn based on deep learning. GEnDDn mainly comprises two steps: First, features of both lncRNAs and diseases are extracted by combining similarity computation, non-negative matrix factorization, and graph attention auto-encoder, respectively. And each lncRNA-disease pair (LDP) is depicted as a vector based on concatenation operation on the extracted features. Subsequently, unknown LDPs are classified by aggregating dual-net neural architecture and deep neural network. Using six different evaluation metrics, we found that GEnDDn surpassed four competing LDA identification methods (SDLDA, LDNFSGB, IPCARF, LDASR) on the lncRNADisease and MNDR databases under fivefold cross-validation experiments on lncRNAs, diseases, LDPs, and independent lncRNAs and independent diseases, respectively. Ablation experiments further validated the powerful LDA prediction performance of GEnDDn. Furthermore, we utilized GEnDDn to find underlying lncRNAs for lung cancer and breast cancer. The results elucidated that there may be dense linkages between IFNG-AS1 and lung cancer as well as between HIF1A-AS1 and breast cancer. The results require further biomedical experimental verification. GEnDDn is publicly available at https://github.com/plhhnu/GEnDDn.

PMID:38733474 | DOI:10.1007/s12539-024-00619-w

Jpn J Radiol. 2024 May 11. doi: 10.1007/s11604-024-01591-7. Online ahead of print.

ABSTRACT

PURPOSE: To compare computed tomography (CT) pulmonary angiography and unenhanced CT to determine the effect of rapid iodine contrast agent infusion on tracheal diameter and lung volume.

MATERIAL AND METHODS: This retrospective study included 101 patients who underwent CT pulmonary angiography and unenhanced CT, for which the time interval between them was within 365 days. CT pulmonary angiography was scanned 20 s after starting the contrast agent injection at the end-inspiratory level. Commercial software, which was developed based on deep learning technique, was used to segment the lung, and its volume was automatically evaluated. The tracheal diameter at the thoracic inlet level was also measured. Then, the ratios for the CT pulmonary angiography to unenhanced CT of the tracheal diameter (TDPAU) and both lung volumes (BLVPAU) were calculated.

RESULTS: Tracheal diameter and both lung volumes were significantly smaller in CT pulmonary angiography (17.2 ± 2.6 mm and 3668 ± 1068 ml, respectively) than those in unenhanced CT (17.7 ± 2.5 mm and 3887 ± 1086 ml, respectively) (p < 0.001 for both). A statistically significant correlation was found between TDPAU and BLVPAU with a correlation coefficient of 0.451 (95% confidence interval, 0.280-0.594) (p < 0.001). No factor showed a significant association with TDPAU. The type of contrast agent had a significant association for BLVPAU (p = 0.042).

CONCLUSIONS: Rapid infusion of iodine contrast agent reduced the tracheal diameter and both lung volumes in CT pulmonary angiography, which was scanned at end-inspiratory level, compared with those in unenhanced CT.

PMID:38733470 | DOI:10.1007/s11604-024-01591-7

J Proteome Res. 2024 May 11. doi: 10.1021/acs.jproteome.3c00367. Online ahead of print.

ABSTRACT

Enzymes are indispensable in many biological processes, and with biomedical literature growing exponentially, effective literature review becomes increasingly challenging. Natural language processing methods offer solutions to streamline this process. This study aims to develop an annotated enzyme corpus for training and evaluating enzyme named entity recognition (NER) models. A novel pipeline, combining dictionary matching and rule-based keyword searching, automatically annotated enzyme entities in >4800 full-text publications. Four deep learning NER models were created with different vocabularies (BioBERT/SciBERT) and architectures (BiLSTM/transformer) and evaluated on 526 manually annotated full-text publications. The annotation pipeline achieved an F1-score of 0.86 (precision = 1.00, recall = 0.76), surpassed by fine-tuned transformers for F1-score (BioBERT: 0.89, SciBERT: 0.88) and recall (0.86) with BiLSTM models having higher precision (0.94) than transformers (0.92). The annotation pipeline runs in seconds on standard laptops with almost perfect precision, but was outperformed by fine-tuned transformers in terms of F1-score and recall, demonstrating generalizability beyond the training data. In comparison, SciBERT-based models exhibited higher precision, and BioBERT-based models exhibited higher recall, highlighting the importance of vocabulary and architecture. These models, representing the first enzyme NER algorithms, enable more effective enzyme text mining and information extraction. Codes for automated annotation and model generation are available from https://github.com/omicsNLP/enzymeNER and https://zenodo.org/doi/10.5281/zenodo.10581586.

PMID:38733346 | DOI:10.1021/acs.jproteome.3c00367

Sensors (Basel). 2024 May 6;24(9):2953. doi: 10.3390/s24092953.

ABSTRACT

Based on the current research on the wine grape variety recognition task, it has been found that traditional deep learning models relying only on a single feature (e.g., fruit or leaf) for classification can face great challenges, especially when there is a high degree of similarity between varieties. In order to effectively distinguish these similar varieties, this study proposes a multisource information fusion method, which is centered on the SynthDiscrim algorithm, aiming to achieve a more comprehensive and accurate wine grape variety recognition. First, this study optimizes and improves the YOLOV7 model and proposes a novel target detection and recognition model called WineYOLO-RAFusion, which significantly improves the fruit localization precision and recognition compared with YOLOV5, YOLOX, and YOLOV7, which are traditional deep learning models. Secondly, building upon the WineYOLO-RAFusion model, this study incorporated the method of multisource information fusion into the model, ultimately forming the MultiFuseYOLO model. Experiments demonstrated that MultiFuseYOLO significantly outperformed other commonly used models in terms of precision, recall, and F1 score, reaching 0.854, 0.815, and 0.833, respectively. Moreover, the method improved the precision of the hard to distinguish Chardonnay and Sauvignon Blanc varieties, which increased the precision from 0.512 to 0.813 for Chardonnay and from 0.533 to 0.775 for Sauvignon Blanc. In conclusion, the MultiFuseYOLO model offers a reliable and comprehensive solution to the task of wine grape variety identification, especially in terms of distinguishing visually similar varieties and realizing high-precision identifications.

PMID:38733058 | DOI:10.3390/s24092953

Sensors (Basel). 2024 May 6;24(9):2948. doi: 10.3390/s24092948.

ABSTRACT

The fetal electrocardiogram (FECG) records changes in the graph of fetal cardiac action potential during conduction, reflecting the developmental status of the fetus in utero and its physiological cardiac activity. Morphological alterations in the FECG can indicate intrauterine hypoxia, fetal distress, and neonatal asphyxia early on, enhancing maternal and fetal safety through prompt clinical intervention, thereby reducing neonatal morbidity and mortality. To reconstruct FECG signals with clear morphological information, this paper proposes a novel deep learning model, CBLS-CycleGAN. The model's generator combines spatial features extracted by the CNN with temporal features extracted by the BiLSTM network, thus ensuring that the reconstructed signals possess combined features with spatial and temporal dependencies. The model's discriminator utilizes PatchGAN, employing small segments of the signal as discriminative inputs to concentrate the training process on capturing signal details. Evaluating the model using two real FECG signal databases, namely "Abdominal and Direct Fetal ECG Database" and "Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeat Annotations", resulted in a mean MSE and MAE of 0.019 and 0.006, respectively. It detects the FQRS compound wave with a sensitivity, positive predictive value, and F1 of 99.51%, 99.57%, and 99.54%, respectively. This paper's model effectively preserves the morphological information of FECG signals, capturing not only the FQRS compound wave but also the fetal P-wave, T-wave, P-R interval, and ST segment information, providing clinicians with crucial diagnostic insights and a scientific foundation for developing rational treatment protocols.

PMID:38733053 | DOI:10.3390/s24092948

Sensors (Basel). 2024 May 6;24(9):2942. doi: 10.3390/s24092942.

ABSTRACT

LiDAR-based 3D object detection and localization are crucial components of autonomous navigation systems, including autonomous vehicles and mobile robots. Most existing LiDAR-based 3D object detection and localization approaches primarily use geometric or structural feature abstractions from LiDAR point clouds. However, these approaches can be susceptible to environmental noise due to adverse weather conditions or the presence of highly scattering media. In this work, we propose an intensity-aware voxel encoder for robust 3D object detection. The proposed voxel encoder generates an intensity histogram that describes the distribution of point intensities within a voxel and is used to enhance the voxel feature set. We integrate this intensity-aware encoder into an efficient single-stage voxel-based detector for 3D object detection. Experimental results obtained using the KITTI dataset show that our method achieves comparable results with respect to the state-of-the-art method for car objects in 3D detection and from a bird's-eye view and superior results for pedestrian and cyclic objects. Furthermore, our model can achieve a detection rate of 40.7 FPS during inference time, which is higher than that of the state-of-the-art methods and incurs a lower computational cost.

PMID:38733047 | DOI:10.3390/s24092942

Sensors (Basel). 2024 May 5;24(9):2937. doi: 10.3390/s24092937.

ABSTRACT

The calculation of land surface temperatures (LSTs) via low-altitude thermal infrared remote (TIR) sensing images at a block scale is gaining attention. However, the accurate calculation of LSTs requires a precise determination of the range of various underlying surfaces in the TIR images, and existing approaches face challenges in effectively segmenting the underlying surfaces in the TIR images. To address this challenge, this study proposes a deep learning (DL) methodology to complete the instance segmentation and quantification of underlying surfaces through the low-altitude TIR image dataset. Mask region-based convolutional neural networks were utilized for pixel-level classification and segmentation with an image dataset of 1350 annotated TIR images of an urban rail transit hub with a complex distribution of underlying surfaces. Subsequently, the hyper-parameters and architecture were optimized for the precise classification of the underlying surfaces. The algorithms were validated using 150 new TIR images, and four evaluation indictors demonstrated that the optimized algorithm outperformed the other algorithms. High-quality segmented masks of the underlying surfaces were generated, and the area of each instance was obtained by counting the true-positive pixels with values of 1. This research promotes the accurate calculation of LSTs based on the low-altitude TIR sensing images.

PMID:38733039 | DOI:10.3390/s24092937