Sci Data. 2024 Sep 27;11(1):1031. doi: 10.1038/s41597-024-03876-y.

ABSTRACT

We present the Manually Annotated GONG Filaments in H-alpha Observations (MAGFiLO v1.0) dataset. This dataset contains 10,244 annotated filaments from 1,593 observations captured by the Global Oscillation Network Group (GONG), spanning the years 2011 through 2022. Each annotation details one filament's segmentation, minimum bounding box, spine, and magnetic field chirality. With a total of over one thousand person-hours of annotation, and a double-blind review process, we ensured high-quality ground-truth data. Our inter-annotator agreement reaches a Kappa score of 0.66. We also verified that the hemispheric preference of filaments as annotated in MAGFiLO aligns with the findings from similar datasets of much smaller sample sizes. MAGFiLO is the first dataset of its size, enabling advanced deep learning models to identify filaments and their features with unprecedented precision. It also provides a testbed for solar physicists interested in large-scale analysis of filaments. In this report, we document the details of the annotation and the post-processing phases that were applied.

PMID:39333537 | DOI:10.1038/s41597-024-03876-y

Sci Data. 2024 Sep 27;11(1):1050. doi: 10.1038/s41597-024-03836-6.

ABSTRACT

Oral cancer is a global health challenge with a difficult histopathological diagnosis. The accurate histopathological interpretation of oral cancer tissue samples remains difficult. However, early diagnosis is very challenging due to a lack of experienced pathologists and inter- observer variability in diagnosis. The application of artificial intelligence (deep learning algorithms) for oral cancer histology images is very promising for rapid diagnosis. However, it requires a quality annotated dataset to build AI models. We present ORCHID (ORal Cancer Histology Image Database), a specialized database generated to advance research in AI-based histology image analytics of oral cancer and precancer. The ORCHID database is an extensive multicenter collection of high-resolution images captured at 1000X effective magnification (100X objective lens), encapsulating various oral cancer and precancer categories, such as oral submucous fibrosis (OSMF) and oral squamous cell carcinoma (OSCC). Additionally, it also contains grade-level sub-classifications for OSCC, such as well- differentiated (WD), moderately-differentiated (MD), and poorly-differentiated (PD). The database seeks to aid in developing innovative artificial intelligence-based rapid diagnostics for OSMF and OSCC, along with subtypes.

PMID:39333529 | DOI:10.1038/s41597-024-03836-6

Clin Transl Oncol. 2024 Sep 27. doi: 10.1007/s12094-024-03717-9. Online ahead of print.

ABSTRACT

OBJECTIVE: The objective of this study was to analyze the incidence and overall survival (OS) of osteosarcoma (OSC) and Ewing's sarcoma (EWS) in a pediatric and adolescent population, employing machine learning (ML) and deep learning (DL) models to predict the likelihood of metastasis.

METHODS: Involving 2465 OSC and 1373 EWS patients aged 0-19 years, from 2004 to 2020. ML techniques-Lasso, Ridge Regression, Elastic Net, and Random Forest-were used alongside a deep learning model based on TensorFlow and Keras, to construct predictive models for metastasis. These models were optimized using grid search with cross-validation and evaluated on their performance metrics, including AUC, sensitivity, and accuracy. The variables' importance in metastasis prediction was determined using SHAP values. Statistical analysis was performed using R software, and an online nomogram was developed for clinical use.

RESULTS: The age-adjusted incidence of OSC and EWS from 2004 to 2020 showed a significant uptrend. The deep learning model, iterated 50 times, outperformed the Random Forest model in both loss and accuracy stabilization. The nomogram created demonstrated accurate survival predictions, as evidenced by its calibration curves and the distinction between high and low-risk groups.

CONCLUSION: The increasing trend in age-adjusted incidence of OSC and EWS highlights the need for continued research and improved therapeutic strategies in this domain. The study employed ML and DL models to predict distant metastasis in pediatric and adolescent patients with OSC and EWS, providing a valuable tool for prognosis. The online nomogram developed as a part of this research enhances the models' clinical utility, offering an accessible means for clinicians to predict survival outcomes effectively.

PMID:39333451 | DOI:10.1007/s12094-024-03717-9

Sports Med Open. 2024 Sep 27;10(1):102. doi: 10.1186/s40798-024-00744-2.

ABSTRACT

BACKGROUND: The growing concern for player safety in rugby has led to an increased focus on head impacts. Previous laboratory studies have shown that rugby headgear significantly reduces peak linear and rotational accelerations compared to no headgear. However, these metrics may have limited relevance in assessing the effectiveness of headgear in preventing strain-based brain injuries like concussions. This study used an instantaneous deep-learning brain injury model to quantify regional brain strain mitigation of rugby headgear during drop tests. Tests were conducted on flat and angled impact surfaces across different heights, using a Hybrid III headform and neck.

RESULTS: Headgear presence generally reduced the peak rotational velocities, with some headgear outperforming others. However, the effect on peak regional brain strains was less consistent. Of the 5 headgear tested, only the newer models that use open cell foams at densities above 45 kg/m3 consistently reduced the peak strain in the cerebrum, corpus callosum, and brainstem. The 3 conventional headgear that use closed cell foams at or below 45 kg/m3 showed no consistent reduction in the peak strain in the cerebrum, corpus callosum, and brainstem.

CONCLUSIONS: The presence of rugby headgear may be able to reduce the severity of head impact exposure during rugby. However, to understand how these findings relate to brain strain mitigation in the field, further investigation into the relationship between the impact conditions in this study and those encountered during actual gameplay is necessary.

PMID:39333426 | DOI:10.1186/s40798-024-00744-2

PLoS One. 2024 Sep 27;19(9):e0303541. doi: 10.1371/journal.pone.0303541. eCollection 2024.

ABSTRACT

Ph-negative Myeloproliferative Neoplasm is a rare yet dangerous disease that can turn into more severe forms of disorders later on. Clinical diagnosis of the disease exists but often requires collecting multiple types of pathologies which can be tedious and time-consuming. Meanwhile, studies on deep learning-based research are rare and often need to rely on a small amount of pathological data due to the rarity of the disease. In addition, the existing research works do not address the data scarcity issue apart from using common techniques like data augmentation, which leaves room for performance improvement. To tackle the issue, the proposed research aims to utilize distilled knowledge learned from a larger dataset to boost the performance of a lightweight model trained on a small MPN dataset. Firstly, a 50-layer ResNet model is trained on a large lymph node image dataset of 3,27,680 images, followed by the trained knowledge being distilled to a small 4-layer CNN model. Afterward, the CNN model is initialized with the pre-trained weights to further train on a small MPN dataset of 300 images. Empirical analysis showcases that the CNN with distilled knowledge achieves 97% accuracy compared to 89.67% accuracy achieved by a clone CNN trained from scratch. The distilled knowledge transfer approach also proves to be more effective than more simple data scarcity handling approaches such as augmentation and manual feature extraction. Overall, the research affirms the effectiveness of transferring distilled knowledge to address the data scarcity issue and achieves better convergence when training on a Ph-Negative MPN image dataset with a lightweight model.

PMID:39331624 | DOI:10.1371/journal.pone.0303541

PLoS One. 2024 Sep 27;19(9):e0310748. doi: 10.1371/journal.pone.0310748. eCollection 2024.

ABSTRACT

Brain tumors are one of the leading diseases imposing a huge morbidity rate across the world every year. Classifying brain tumors accurately plays a crucial role in clinical diagnosis and improves the overall healthcare process. ML techniques have shown promise in accurately classifying brain tumors based on medical imaging data such as MRI scans. These techniques aid in detecting and planning treatment early, improving patient outcomes. However, medical image datasets are frequently affected by a significant class imbalance, especially when benign tumors outnumber malignant tumors in number. This study presents an explainable ensemble-based pipeline for brain tumor classification that integrates a Dual-GAN mechanism with feature extraction techniques, specifically designed for highly imbalanced data. This Dual-GAN mechanism facilitates the generation of synthetic minority class samples, addressing the class imbalance issue without compromising the original quality of the data. Additionally, the integration of different feature extraction methods facilitates capturing precise and informative features. This study proposes a novel deep ensemble feature extraction (DeepEFE) framework that surpasses other benchmark ML and deep learning models with an accuracy of 98.15%. This study focuses on achieving high classification accuracy while prioritizing stable performance. By incorporating Grad-CAM, it enhances the transparency and interpretability of the overall classification process. This research identifies the most relevant and contributing parts of the input images toward accurate outcomes enhancing the reliability of the proposed pipeline. The significantly improved Precision, Sensitivity and F1-Score demonstrate the effectiveness of the proposed mechanism in handling class imbalance and improving the overall accuracy. Furthermore, the integration of explainability enhances the transparency of the classification process to establish a reliable model for brain tumor classification, encouraging their adoption in clinical practice promoting trust in decision-making processes.

PMID:39331600 | DOI:10.1371/journal.pone.0310748

Bioinformatics. 2024 Sep 27:btae581. doi: 10.1093/bioinformatics/btae581. Online ahead of print.

ABSTRACT

MOTIVATION: Nucleotide-binding leucine-rich repeat (NLR) family is a class of immune receptors capable of detecting and defending against pathogen invasion. They have been widely used in crop breeding. Notably, the correspondence between NLRs and effectors (CNE) determines the applicability and effectiveness of NLRs. Unfortunately, CNE data is very scarce. In fact, we've found a substantial 91,291 NLRs confirmed via wet experiments and bioinformatics methods but only 387 CNEs are recognized, which greatly restricts the potential application of NLRs.

RESULTS: We propose a deep learning algorithm called ProNEP to identify NLR-effector pairs in a high-throughput manner. Specifically, we conceptualized the CNE prediction task as a protein-protein interaction (PPI) prediction task. Then, ProNEP predicts the interaction between NLRs and effectors by combining the transfer learning with a bilinear attention network. ProNEP achieves superior performance against state-of-the-art models designed for PPI predictions. Based on ProNEP, we conduct extensive identification of potential CNEs for 91,291 NLRs. With the rapid accumulation of genomic data, we expect that this tool will be widely used to predict CNEs in new species, advancing biology, immunology, and breeding.

AVAILABILITY: The ProNEP is available at http://nerrd.cn/#/prediction. The project code is available at https://github.com/ QiaoYJYJ/ProNEP.

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:39331576 | DOI:10.1093/bioinformatics/btae581

IEEE J Biomed Health Inform. 2024 Sep 27;PP. doi: 10.1109/JBHI.2024.3470310. Online ahead of print.

ABSTRACT

Step width is vital for gait stability, postural balance control, and fall risk reduction. However, estimating step width typically requires either fixed cameras or a full kinematic body suit of wearable inertial measurement units (IMUs), both of which are often too expensive and time-consuming for clinical application. We thus propose a novel data-augmented deep learning model for estimating step width in individuals with and without neurodegenerative disease using a minimal set of wearable IMUs. Twelve patients with neurodegenerative, clinically diagnosed Spinocerebellar ataxia type 3 (SCA3) performed over ground walking trials, and seventeen healthy individuals performed treadmill walking trials at various speeds and gait modifications while wearing IMUs on each shank and the pelvis. Results demonstrated step width mean absolute errors of 3.3 ± 0.7cm and 2.9 ± 0.5cm for the neurodegenerative and healthy groups, respectively, which were below the minimal clinically important difference of 6.0cm. Step width variability mean absolute errors were 1.5cm and 0.8cm for neurodegenerative and healthy groups, respectively. Data augmentation significantly improved accuracy performance in the neurodegenerative group, likely because they exhibited larger variations in walking kinematics as compared with healthy subjects. These results could enable clinically meaningful and accurate portable step width monitoring for individuals with and without neurodegenerative disease, potentially enhancing rehabilitative training, assessment, and dynamic balance control in clinical and real-life settings.

PMID:39331558 | DOI:10.1109/JBHI.2024.3470310

IEEE Trans Pattern Anal Mach Intell. 2024 Sep 27;PP. doi: 10.1109/TPAMI.2024.3469952. Online ahead of print.

ABSTRACT

Deep learning models have emerged as strong and efficient tools that can be applied to a broad spectrum of complex learning problems and many real-world applications. However, more and more works show that deep models are vulnerable to adversarial examples. Compared to vanilla attack settings, this paper advocates a more practical setting of data-free black-box attack, for which the attackers can completely not access the structures and parameters of the target model, as well as the intermediate features and any training data associated with the model. To tackle this task, previous methods generate transferable adversarial examples from a transparent substitute model to the target model. However, we found that these works have the limitations of taking static substitute model structure for different targets, only using hard synthesized examples once, and still relying on data statistics of the target model. This may potentially harm the performance of attacking the target model. To this end, we propose a novel Dynamic Routing and Knowledge Re-Learning framework (DraKe) to effectively learn a dynamic substitute model from the target model. Specifically, given synthesized training samples, a dynamic substitute structure learning strategy is proposed to adaptively generate optimal substitute model structure via a policy network according to different target models and tasks. To facilitate the substitute training, we present a graph-based structure information learning to capture the structural knowledge learned from the target model. For the inherent limitation that online data generation can only be learned once, a dynamic knowledge re-learning strategy is proposed to adjust the weights of optimization objectives and re-learn hard samples. Extensive experiments on four public image classification datasets and one face recognition benchmark are conducted to evaluate the efficacy of our Drake. We can obtain significant improvement compared with state-of-the-art competitors. More importantly, our DraKe consistently achieves attack superiority for different target models (e.g., residual networks, and vision transformers), showing great potential for complex real-world applications.

PMID:39331554 | DOI:10.1109/TPAMI.2024.3469952

IEEE/ACM Trans Comput Biol Bioinform. 2024 Sep 27;PP. doi: 10.1109/TCBB.2024.3468434. Online ahead of print.

ABSTRACT

Drug Target Interaction (DTI) prediction plays a crucial role in in-silico drug discovery, especially for deep learning (DL) models. Along this line, existing methods usually first extract features from drugs and target proteins, and use drug-target pairs to train DL models. However, these DL-based methods essentially rely on similar structures and patterns defined by the homologous proteins from a large amount of data. When few drug-target interactions are known for a newly discovered protein and its homologous proteins, prediction performance can suffer notable reduction. In this paper, we propose a novel Protein-Context enhanced Master/Slave Framework (PCMS), for zero-shot DTI prediction. This framework facilitates the efficient discovery of ligands for newly discovered target proteins, addressing the challenge of predicting interactions without prior data. Specifically, the PCMS framework consists of two main components: a Master Learner and a Slave Learner. The Master Learner first learns the target protein context information, and then adaptively generates the corresponding parameters for the Slave Learner. The Slave Learner then perform zero-shot DTI prediction in different protein contexts. Extensive experiments verify the effectiveness of our PCMS compared to state-of-the-art methods in various metrics on two public datasets. The Code and the processed Data will be open once the paper is accepted.

PMID:39331551 | DOI:10.1109/TCBB.2024.3468434

IEEE Trans Biomed Eng. 2024 Sep 27;PP. doi: 10.1109/TBME.2024.3470534. Online ahead of print.

ABSTRACT

BACKGROUND: Sleep staging is critical for diagnosing sleep disorders. Traditional methods in clinical settings involve time-intensive scoring procedures. Recent advancements in data-driven algorithms using photoplethysmogram (PPG) time series have shown promise in automating sleep staging in adults. However, for children, algorithm development is hindered by the limited availability of datasets, with the Childhood Adenotonsillectomy Trial (CHAT) being the only substantial source, comprising recordings from children aged 5-10. This limitation constrains the evaluation of algorithmic generalization performance.

METHODS: We employed a deep learning model for sleep staging from PPG, initially trained using a large dataset of adult sleep recordings, and fine-tuned it on 80% of the CHAT dataset (CHAT-train) for the task of three-class sleep staging (wake, REM, non-REM). The resulting algorithm performance was compared to the same model architecture but trained from scratch on CHAT-train (benchmark). The algorithms are evaluated on the local test set, denoted CHAT-test, as well as on a newly introduced independent dataset.

RESULTS: Our deep learning algorithm achieved a Cohen's Kappa of 0.88 on CHAT-test (versus 0.65), and demonstrated generalization capabilities with a Kappa of 0.72 on the external Ichilov dataset for children above 5 years old (versus 0.64) and 0.64 for those below 5 (versus 0.53).

SIGNIFICANCE: This research establishes a new state-of-the-art performance for the task of sleep staging in children using raw PPG. The findings underscore the value of transfer learning from the adults to children domain. However, the reduced performance in children under 5 suggests the need for further research and additional datasets covering a broader pediatric age range to fully address generalization limitations.

PMID:39331540 | DOI:10.1109/TBME.2024.3470534

Environ Monit Assess. 2024 Sep 27;196(10):983. doi: 10.1007/s10661-024-13101-3.

ABSTRACT

Some recent studies highlight that vehicular traffic and honking contribute to more than 50% of noise pollution in urban or sub-urban areas in developing countries, including Indian cities. Frequent honking has an adverse effect on health and hampers road safety, the environment, etc. Therefore, recognizing the various vehicle honks and classifying the honk of different vehicles can provide good insights into environmental noise pollution. Moreover, classifying honks based on vehicle types allows for the inference of contextual information of a location, area, or traffic. So far, the researchers have done outdoor sound classification and honk detection, where vehicular honks are collected in a controlled environment or in the absence of ambient noise. Such classification models fail to classify honk based on vehicle types. Therefore, it becomes imperative to design a system that can detect and classify honks of different types of vehicles to infer some contextual information. This paper presents a novel framework A C lassi H onk that performs raw vehicular honk sensing, data labeling, and classifies the honk into three major groups, i.e., light-weight vehicles, medium-weight vehicles, and heavy-weight vehicles. Raw audio samples of different vehicular honking are collected based on spatio-temporal characteristics and converted them into spectrogram images. A deep learning-based multi-label autoencoder model (MAE) is proposed for automated labeling of the unlabeled data samples, which provides 97.64% accuracy in contrast to existing deep learning-based data labeling methods. Further, various pre-trained models, namely Inception V3, ResNet50, MobileNet, and ShuffleNet are used and proposed an Ensembled Transfer Learning model (EnTL) for vehicle honks classification and performed comparative analysis. Results reveal that EnTL exhibits the best performance compared to pre-trained models and achieves 96.72% accuracy in our dataset. In addition, context of a location is identified based on these classified honk signatures in a city.

PMID:39331183 | DOI:10.1007/s10661-024-13101-3

Breast Dis. 2024;43(1):257-270. doi: 10.3233/BD-240018.

ABSTRACT

Breast Cancer is the leading form of cancer found in women and a major cause of increased mortality rates among them. However, manual diagnosis of the disease is time-consuming and often limited by the availability of screening systems. Thus, there is a pressing need for an automatic diagnosis system that can quickly detect cancer in its early stages. Data mining and machine learning techniques have emerged as valuable tools in developing such a system. In this study we investigated the performance of several machine learning models on the Wisconsin Breast Cancer (original) dataset with a particular emphasis on finding which models perform the best for breast cancer diagnosis. The study also explores the contrast between the proposed ANN methodology and conventional machine learning techniques. The comparison between the methods employed in the current study and those utilized in earlier research on the Wisconsin Breast Cancer dataset is also compared. The findings of this study are in line with those of previous studies which also highlighted the efficacy of SVM, Decision Tree, CART, ANN, and ELM ANN for breast cancer detection. Several classifiers achieved high accuracy, precision and F1 scores for benign and malignant tumours, respectively. It is also found that models with hyperparameter adjustment performed better than those without and boosting methods like as XGBoost, Adaboost, and Gradient Boost consistently performed well across benign and malignant tumours. The study emphasizes the significance of hyperparameter tuning and the efficacy of boosting algorithms in addressing the complexity and nonlinearity of data. Using the Wisconsin Breast Cancer (original) dataset, a detailed summary of the current status of research on breast cancer diagnosis is provided.

PMID:39331085 | DOI:10.3233/BD-240018

Vet Sci. 2024 Sep 1;11(9):399. doi: 10.3390/vetsci11090399.

ABSTRACT

Body condition score (BCS) is a common tool used to assess the welfare of dairy cows and is based on scoring animals according to their external appearance. If the BCS of dairy cows deviates from the required value, it can lead to diseases caused by metabolic problems in the animal, increased medication costs, low productivity, and even the loss of dairy cows. BCS scores for dairy cows on farms are mostly determined by observation based on expert knowledge and experience. This study proposes an automatic classification system for BCS determination in dairy cows using the YOLOv8x deep learning architecture. In this study, firstly, an original dataset was prepared by dividing the BCS scale into five different classes of Emaciated, Poor, Good, Fat, and Obese for images of Holstein and Simmental cow breeds collected from different farms. In the experimental analyses performed on the dataset prepared in this study, the BCS values of 102 out of a total of 126 cow images in the test set were correctly classified using the proposed YOLOv8x deep learning architecture. Furthermore, an average accuracy of 0.81 was achieved for all BCS classes in Holstein and Simmental cows. In addition, the average area under the precision-recall curve was 0.87. In conclusion, the BCS classification system for dairy cows proposed in this study may allow for the accurate observation of animals with rapid declines in body condition. In addition, the BCS classification system can be used as a tool for production decision-makers in early lactation to reduce the negative energy balance.

PMID:39330779 | DOI:10.3390/vetsci11090399

Tomography. 2024 Sep 13;10(9):1513-1526. doi: 10.3390/tomography10090111.

ABSTRACT

BACKGROUND: The cross-sectional area of skeletal muscles at the level of the third lumbar vertebra (L3) measured from computed tomography (CT) images is an established imaging biomarker used to assess patients' nutritional status. With the increasing prevalence of low-dose CT scans in clinical practice, accurate and automated skeletal muscle segmentation at the L3 level in low-dose CT images has become an issue to address. This study proposed a lightweight algorithm for automated segmentation of skeletal muscles at the L3 level in low-dose CT images.

METHODS: This study included 57 patients with rectal cancer, with both low-dose plain and contrast-enhanced pelvic CT image series acquired using a radiotherapy CT scanner. A training set of 30 randomly selected patients was used to develop a lightweight segmentation algorithm, and the other 27 patients were used as the test set. A radiologist selected the most representative axial CT image at the L3 level for both the image series for all the patients, and three groups of observers manually annotated the skeletal muscles in the 54 CT images of the test set as the gold standard. The performance of the proposed algorithm was evaluated in terms of the Dice similarity coefficient (DSC), precision, recall, 95th percentile of the Hausdorff distance (HD95), and average surface distance (ASD). The running time of the proposed algorithm was recorded. An open source deep learning-based AutoMATICA algorithm was compared with the proposed algorithm. The inter-observer variations were also used as the reference.

RESULTS: The DSC, precision, recall, HD95, ASD, and running time were 93.2 ± 1.9% (mean ± standard deviation), 96.7 ± 2.9%, 90.0 ± 2.9%, 4.8 ± 1.3 mm, 0.8 ± 0.2 mm, and 303 ± 43 ms (on CPU) for the proposed algorithm, and 94.1 ± 4.1%, 92.7 ± 5.5%, 95.7 ± 4.0%, 7.4 ± 5.7 mm, 0.9 ± 0.6 mm, and 448 ± 40 ms (on GPU) for AutoMATICA, respectively. The differences between the proposed algorithm and the inter-observer reference were 4.7%, 1.2%, 7.9%, 3.2 mm, and 0.6 mm, respectively, for the averaged DSC, precision, recall, HD95, and ASD.

CONCLUSION: The proposed algorithm can be used to segment skeletal muscles at the L3 level in either the plain or enhanced low-dose CT images.

PMID:39330757 | DOI:10.3390/tomography10090111

Tomography. 2024 Sep 5;10(9):1488-1500. doi: 10.3390/tomography10090109.

ABSTRACT

Pneumonia caused by novel coronavirus is an acute respiratory infectious disease. Its rapid spread in a short period of time has brought great challenges for global public health. The use of deep learning and radiomics methods can effectively distinguish the subtypes of lung diseases, provide better clinical prognosis accuracy, and assist clinicians, enabling them to adjust the clinical management level in time. The main goal of this study is to verify the performance of deep learning and radiomics methods in the classification of COVID-19 lesions and reveal the image characteristics of COVID-19 lung disease. An MFPN neural network model was proposed to extract the depth features of lesions, and six machine-learning methods were used to compare the classification performance of deep features, key radiomics features and combined features for COVID-19 lung lesions. The results show that in the COVID-19 image classification task, the classification method combining radiomics and deep features can achieve good classification results and has certain clinical application value.

PMID:39330755 | DOI:10.3390/tomography10090109

Tomography. 2024 Sep 1;10(9):1397-1410. doi: 10.3390/tomography10090105.

ABSTRACT

The Brain Tumor Segmentation (BraTS) Challenge has been a main driver of the development of deep learning (DL) algorithms and provides by far the largest publicly available expert-annotated brain tumour dataset but contains solely preoperative examinations. The aim of our study was to facilitate the use of the BraTS dataset for training DL brain tumour segmentation algorithms for a postoperative setting. To this end, we introduced an automatic conversion of the three-label BraTS annotation protocol to a two-label annotation protocol suitable for postoperative brain tumour segmentation. To assess the viability of the label conversion, we trained a DL algorithm using both the three-label and the two-label annotation protocols. We assessed the models pre- and postoperatively and compared the performance with a state-of-the-art DL method. The DL algorithm trained using the BraTS three-label annotation misclassified parts of 10 out of 41 fluid-filled resection cavities in 72 postoperative glioblastoma MRIs, whereas the two-label model showed no such inaccuracies. The tumour segmentation performance of the two-label model both pre- and postoperatively was comparable to that of a state-of-the-art algorithm for tumour volumes larger than 1 cm3. Our study enables using the BraTS dataset as a basis for the training of DL algorithms for postoperative tumour segmentation.

PMID:39330751 | DOI:10.3390/tomography10090105

Nanomaterials (Basel). 2024 Sep 14;14(18):1497. doi: 10.3390/nano14181497.

ABSTRACT

The correlation between metasurface structures and their corresponding absorption spectra is inherently complex due to intricate physical interactions. Additionally, the reliance on Maxwell's equations for simulating these relationships leads to extensive computational demands, significantly hindering rapid development in this area. Numerous researchers have employed artificial intelligence (AI) models to predict absorption spectra. However, these models often act as black boxes. Despite training high-performance models, it remains challenging to verify if they are fitting to rational patterns or merely guessing outcomes. To address these challenges, we introduce the Explainable Encoder-Prediction-Reconstruction (EEPR) framework, which separates the prediction process into feature extraction and spectra generation, facilitating a deeper understanding of the physical relationships between metasurface structures and spectra and unveiling the model's operations at the feature level. Our model achieves a 66.23% reduction in average Mean Square Error (MSE), with an MSE of 2.843 × 10-4 compared to the average MSE of 8.421×10-4 for mainstream networks. Additionally, our model operates approximately 500,000 times faster than traditional simulations based on Maxwell's equations, with a time of 3×10-3 seconds per sample, and demonstrates excellent generalization capabilities. By utilizing the EEPR framework, we achieve feature-level explainability and offer insights into the physical properties and their impact on metasurface structures, going beyond the pixel-level explanations provided by existing research. Additionally, we demonstrate the capability to adjust absorption by changing the metasurface at the feature level. These insights potentially empower designers to refine structures and enhance their trust in AI applications.

PMID:39330654 | DOI:10.3390/nano14181497

J Imaging. 2024 Sep 23;10(9):237. doi: 10.3390/jimaging10090237.

ABSTRACT

Point-of-gaze estimation is part of a larger set of tasks aimed at improving user experience, providing business insights, or facilitating interactions with different devices. There has been a growing interest in this task, particularly due to the need for upgrades in e-meeting platforms during the pandemic when on-site activities were no longer possible for educational institutions, corporations, and other organizations. Current research advancements are focusing on more complex methodologies for data collection and task implementation, creating a gap that we intend to address with our contributions. Thus, we introduce a methodology for data acquisition that shows promise due to its nonrestrictive and straightforward nature, notably increasing the yield of collected data without compromising diversity or quality. Additionally, we present a novel and efficient convolutional neural network specifically tailored for calibration-free point-of-gaze estimation that outperforms current state-of-the-art methods on the MPIIFaceGaze dataset by a substantial margin, and sets a strong baseline on our own data.

PMID:39330457 | DOI:10.3390/jimaging10090237

J Imaging. 2024 Sep 20;10(9):234. doi: 10.3390/jimaging10090234.

ABSTRACT

In the computer-aided diagnosis of lung cancer, the automatic segmentation of pulmonary nodules and the classification of benign and malignant tumors are two fundamental tasks. However, deep learning models often overlook the potential benefits of task correlations in improving their respective performances, as they are typically designed for a single task only. Therefore, we propose a multi-task network (MT-Net) that integrates shared backbone architecture and a prediction distillation structure for the simultaneous segmentation and classification of pulmonary nodules. The model comprises a coarse segmentation subnetwork (Coarse Seg-net), a cooperative classification subnetwork (Class-net), and a cooperative segmentation subnetwork (Fine Seg-net). Coarse Seg-net and Fine Seg-net share identical structure, where Coarse Seg-net provides prior location information for the subsequent Fine Seg-net and Class-net, thereby boosting pulmonary nodule segmentation and classification performance. We quantitatively and qualitatively analyzed the performance of the model by using the public dataset LIDC-IDRI. Our results show that the model achieves a Dice similarity coefficient (DI) index of 83.2% for pulmonary nodule segmentation, as well as an accuracy (ACC) of 91.9% for benign and malignant pulmonary nodule classification, which is competitive with other state-of-the-art methods. The experimental results demonstrate that the performance of pulmonary nodule segmentation and classification can be improved by a unified model that leverages the potential correlation between tasks.

PMID:39330454 | DOI:10.3390/jimaging10090234