RESUMO
Semi-supervised medical image segmentation (SSMIS) has witnessed substantial advancements by leveraging limited labeled data and abundant unlabeled data. Nevertheless, existing state-of-the-art (SOTA) methods encounter challenges in accurately predicting labels for the unlabeled data, giving rise to disruptive noise during training and susceptibility to erroneous information overfitting. Moreover, applying perturbations to inaccurate predictions further impedes consistent learning. To address these concerns, we propose a novel cross-head mutual mean-teaching network (CMMT-Net) incorporated weak-strong data augmentations, thereby benefiting both co-training and consistency learning. More concretely, our CMMT-Net extends the cross-head co-training paradigm by introducing two auxiliary mean teacher models, which yield more accurate predictions and provide supplementary supervision. The predictions derived from weakly augmented samples generated by one mean teacher are leveraged to guide the training of another student with strongly augmented samples. Furthermore, two distinct yet synergistic data perturbations at the pixel and region levels are introduced. We propose mutual virtual adversarial training (MVAT) to smooth the decision boundary and enhance feature representations, and a cross-set CutMix strategy to generate more diverse training samples for capturing inherent structural data information. Notably, CMMT-Net simultaneously implements data, feature, and network perturbations, amplifying model diversity and generalization performance. Experimental results on three publicly available datasets indicate that our approach yields remarkable improvements over previous SOTA methods across various semi-supervised scenarios. The code is available at https://github.com/Leesoon1984/CMMT-Net.
RESUMO
Non-intrusive load monitoring (NILM) can obtain fine-grained power consumption information for individual appliances within the user without installing additional hardware sensors. With the rapid development of the deep learning model, many methods have been utilized to address NILM problems and have achieved enhanced appliance identification performance. However, supervised learning models require a substantial volume of annotated data to function effectively, which is time-consuming, laborious, and difficult to implement in real scenarios. In this paper, we propose a novel semi-supervised learning method that combines consistency regularization and pseudo-labels to help identification of appliances with limited labeled data and an abundance of unlabeled data. In addition, given the different learning difficulties of various appliance categories, for example, feature learning is more difficult for multi-state appliances than two-state appliances, the thresholds employed for different appliances are adjusted in a flexible way at each time step so that the informative unlabeled data and their pseudo-labels can be delivered. Experiments have been conducted on publicly available datasets, and the results indicate that the proposed method attains superior appliance identification performance compared to cutting-edge methods.
RESUMO
Background: Three-dimensional cephalometric analysis is crucial in craniomaxillofacial assessment, with landmarks detection in craniomaxillofacial (CMF) CT scans being a key component. However, creating robust deep learning models for this task typically requires extensive CMF CT datasets annotated by experienced medical professionals, a process that is time-consuming and labor-intensive. Conversely, acquiring large volume of unlabeled CMF CT data is relatively straightforward. Thus, semi-supervised learning (SSL), leveraging limited labeled data supplemented by sufficient unlabeled dataset, could be a viable solution to this challenge. Method: We developed an SSL model, named CephaloMatch, based on a strong-weak perturbation consistency framework. The proposed SSL model incorporates a head position rectification technique through coarse detection to enhance consistency between labeled and unlabeled datasets and a multilayers perturbation method which is employed to expand the perturbation space. The proposed SSL model was assessed using 362 CMF CT scans, divided into a training set (60 scans), a validation set (14 scans), and an unlabeled set (288 scans). Result: The proposed SSL model attained a detection error of 1.60 ± 0.87 mm, significantly surpassing the performance of conventional fully supervised learning model (1.94 ± 1.12 mm). Notably, the proposed SSL model achieved equivalent detection accuracy (1.91 ± 1.00 mm) with only half the labeled dataset, compared to the fully supervised learning model. Conclusions: The proposed SSL model demonstrated exceptional performance in landmarks detection using a limited labeled CMF CT dataset, significantly reducing the workload of medical professionals and enhances the accuracy of 3D cephalometric analysis.
RESUMO
Electronic Health Records (EHRs) play a crucial role in shaping predictive are models, yet they encounter challenges such as significant data gaps and class imbalances. Traditional Graph Neural Network (GNN) approaches have limitations in fully leveraging neighbourhood data or demanding intensive computational requirements for regularisation. To address this challenge, we introduce CliqueFluxNet, a novel framework that innovatively constructs a patient similarity graph to maximise cliques, thereby highlighting strong inter-patient connections. At the heart of CliqueFluxNet lies its stochastic edge fluxing strategy - a dynamic process involving random edge addition and removal during training. This strategy aims to enhance the model's generalisability and mitigate overfitting. Our empirical analysis, conducted on MIMIC-III and eICU datasets, focuses on the tasks of mortality and readmission prediction. It demonstrates significant progress in representation learning, particularly in scenarios with limited data availability. Qualitative assessments further underscore CliqueFluxNet's effectiveness in extracting meaningful EHR representations, solidifying its potential for advancing GNN applications in healthcare analytics.
RESUMO
Human action recognition (HAR) technology based on radar signals has garnered significant attention from both industry and academia due to its exceptional privacy-preserving capabilities, noncontact sensing characteristics, and insensitivity to lighting conditions. However, the scarcity of accurately labeled human radar data poses a significant challenge in meeting the demand for large-scale training datasets required by deep model-based HAR technology, thus substantially impeding technological advancements in this field. To address this issue, a semi-supervised learning algorithm, MF-Match, is proposed in this paper. This algorithm computes pseudo-labels for larger-scale unsupervised radar data, enabling the model to extract embedded human behavioral information and enhance the accuracy of HAR algorithms. Furthermore, the method incorporates contrastive learning principles to improve the quality of model-generated pseudo-labels and mitigate the impact of mislabeled pseudo-labels on recognition performance. Experimental results demonstrate that this method achieves action recognition accuracies of 86.69% and 91.48% on two widely used radar spectrum datasets, respectively, utilizing only 10% labeled data, thereby validating the effectiveness of the proposed approach.
Assuntos
Algoritmos , Humanos , Radar , Aprendizado de Máquina Supervisionado , Reconhecimento Automatizado de Padrão/métodos , Atividades HumanasRESUMO
Deep neural networks are commonly used for automated medical image segmentation, but models will frequently struggle to generalize well across different imaging modalities. This issue is particularly problematic due to the limited availability of annotated data, both in the target as well as the source modality, making it difficult to deploy these models on a larger scale. To overcome these challenges, we propose a new semi-supervised training strategy called MoDATTS. Our approach is designed for accurate cross-modality 3D tumor segmentation on unpaired bi-modal datasets. An image-to-image translation strategy between modalities is used to produce synthetic but annotated images and labels in the desired modality and improve generalization to the unannotated target modality. We also use powerful vision transformer architectures for both image translation (TransUNet) and segmentation (Medformer) tasks and introduce an iterative self-training procedure in the later task to further close the domain gap between modalities, thus also training on unlabeled images in the target modality. MoDATTS additionally allows the possibility to exploit image-level labels with a semi-supervised objective that encourages the model to disentangle tumors from the background. This semi-supervised methodology helps in particular to maintain downstream segmentation performance when pixel-level label scarcity is also present in the source modality dataset, or when the source dataset contains healthy controls. The proposed model achieves superior performance compared to other methods from participating teams in the CrossMoDA 2022 vestibular schwannoma (VS) segmentation challenge, as evidenced by its reported top Dice score of 0.87±0.04 for the VS segmentation. MoDATTS also yields consistent improvements in Dice scores over baselines on a cross-modality adult brain gliomas segmentation task composed of four different contrasts from the BraTS 2020 challenge dataset, where 95% of a target supervised model performance is reached when no target modality annotations are available. We report that 99% and 100% of this maximum performance can be attained if 20% and 50% of the target data is additionally annotated, which further demonstrates that MoDATTS can be leveraged to reduce the annotation burden.
RESUMO
Purpose: Synthetic datasets hold the potential to offer cost-effective alternatives to clinical data, ensuring privacy protections and potentially addressing biases in clinical data. We present a method leveraging such datasets to train a machine learning algorithm applied as part of a computer-aided detection (CADe) system. Approach: Our proposed approach utilizes clinically acquired computed tomography (CT) scans of a physical anthropomorphic phantom into which manufactured lesions were inserted to train a machine learning algorithm. We treated the training database obtained from the anthropomorphic phantom as a simplified representation of clinical data and increased the variability in this dataset using a set of randomized and parameterized augmentations. Furthermore, to mitigate the inherent differences between phantom and clinical datasets, we investigated adding unlabeled clinical data into the training pipeline. Results: We apply our proposed method to the false positive reduction stage of a lung nodule CADe system in CT scans, in which regions of interest containing potential lesions are classified as nodule or non-nodule regions. Experimental results demonstrate the effectiveness of the proposed method; the system trained on labeled data from physical phantom scans and unlabeled clinical data achieves a sensitivity of 90% at eight false positives per scan. Furthermore, the experimental results demonstrate the benefit of the physical phantom in which the performance in terms of competitive performance metric increased by 6% when a training set consisting of 50 clinical CT scans was enlarged by the scans obtained from the physical phantom. Conclusions: The scalability of synthetic datasets can lead to improved CADe performance, particularly in scenarios in which the size of the labeled clinical data is limited or subject to inherent bias. Our proposed approach demonstrates an effective utilization of synthetic datasets for training machine learning algorithms.
RESUMO
PURPOSE: In recent years, the use of deep learning for medical image segmentation has become a popular trend, but its development also faces some challenges. Firstly, due to the specialized nature of medical data, precise annotation is time-consuming and labor-intensive. Training neural networks effectively with limited labeled data is a significant challenge in medical image analysis. Secondly, convolutional neural networks commonly used for medical image segmentation research often focus on local features in images. However, the recognition of complex anatomical structures or irregular lesions often requires the assistance of both local and global information, which has led to a bottleneck in its development. Addressing these two issues, in this paper, we propose a novel network architecture. METHODS: We integrate a shift window mechanism to learn more comprehensive semantic information and employ a semi-supervised learning strategy by incorporating a flexible amount of unlabeled data. Specifically, a typical U-shaped encoder-decoder structure is applied to obtain rich feature maps. Each encoder is designed as a dual-branch structure, containing Swin modules equipped with windows of different size to capture features of multiple scales. To effectively utilize unlabeled data, a level set function is introduced to establish consistency between the function regression and pixel classification. RESULTS: We conducted experiments on the COVID-19 CT dataset and DRIVE dataset and compared our approach with various semi-supervised and fully supervised learning models. On the COVID-19 CT dataset, we achieved a segmentation accuracy of up to 74.56%. Our segmentation accuracy on the DRIVE dataset was 79.79%. CONCLUSIONS: The results demonstrate the outstanding performance of our method on several commonly used evaluation metrics. The high segmentation accuracy of our model demonstrates that utilizing Swin modules with different window sizes can enhance the feature extraction capability of the model, and the level set function can enable semi-supervised models to more effectively utilize unlabeled data. This provides meaningful insights for the application of deep learning in medical image segmentation. Our code will be released once the manuscript is accepted for publication.
RESUMO
Materials characterization using electron backscatter diffraction (EBSD) requires indexing the orientation of the measured region from Kikuchi patterns. The quality of Kikuchi patterns can degrade due to pattern overlaps arising from two or more orientations, in the presence of defects or grain boundaries. In this work, we employ constrained nonnegative matrix factorization to segment a microstructure with small grain misorientations, (<1∘), and predict the amount of pattern overlap. First, we implement the method on mixed simulated patterns-that replicates a pattern overlap scenario, and demonstrate the resolution limit of pattern mixing or factorization resolution using a weight metric. Subsequently, we segment a single-crystal dendritic microstructure and compare the results with high-resolution EBSD. By utilizing weight metrics across a low-angle grain boundary, we demonstrate how very small misorientations/low-angle grain boundaries can be resolved at a pixel level. Our approach constitutes a versatile and robust tool, complementing other fast indexing methods for microstructure characterization.
RESUMO
Particulate matter (PM) is defined by the Texas Commission on Environmental Quality (TCEQ) as "a mixture of solid particles and liquid droplets found in the air". These particles vary widely in size. Those particles that are less than 2.5 µm in aerodynamic diameter are known as Particulate Matter 2.5 or PM2.5. Urban haze pollution represented by PM2.5 is becoming serious, so air pollution monitoring is very important. However, due to high cost, the number of air monitoring stations is limited. Our work focuses on integrating multi-source heterogeneous data of Nanchang, China, which includes Taxi track, human mobility, Road networks, Points of Interest (POIs), Meteorology (e.g., temperature, dew point, humidity, wind speed, wind direction, atmospheric pressure, weather activity, weather conditions) and PM2.5 forecast data of air monitoring stations. This research presents an innovative approach to air quality prediction by integrating the above data sets from various sources and utilizing diverse architectures in Nanchang City, China. So for that, semi-supervised learning techniques will be used, namely collaborative training algorithm Co-Training (Co-T), who further adjusting algorithm Tri-Training (Tri-T). The objective is to accurately estimate haze pollution by integrating and using these multi-source heterogeneous data. We achieved this for the first time by employing a semi-supervised co-training strategy to accurately estimate pollution levels after applying the U-air system to environmental data. In particular, the algorithm of U-Air system is reproduced on these highly diverse heterogeneous data of Nanchang City, and the semi-supervised learning Co-T and Tri-T are used to conduct more detailed urban haze pollution prediction. Compared with Co-T, which train time classifier (TC) and subspace classifier (SC) respectively from the separated spatio-temporal perspective, the Tri-T is more accurate with a and faster because of its testing accuracy up to 85.62 %. The forecast results also present the potential of the city multi-source heterogeneous data and the effectiveness of the semi-supervised learning. We hope that this synthesis will motivate atmospheric environmental officials, scientists, and environmentalists in China to explore machine learning technology for controlling the discharge of pollutants and environmental management.
RESUMO
Cardiotocography (CTG) measurements are critical for assessing fetal wellbeing during monitoring, and accurate assessment requires well-traceable CTG signals. The current FHR calculation algorithm, based on autocorrelation to Doppler ultrasound (DUS) signals, often results in periods of loss owing to its inability to differentiate signals. We hypothesized that classifying DUS signals by type could be a solution and proposed that an artificial intelligence (AI)-based approach could be used for classification. However, limited studies have incorporated the use of AI for DUS signals because of the limited data availability. Therefore, this study focused on evaluating the effectiveness of semi-supervised learning in enhancing classification accuracy, even in limited datasets, for DUS signals. Data comprising fetal heartbeat, artifacts, and two other categories were created from non-stress tests and labor DUS signals. With labeled and unlabeled data totaling 9,600 and 48,000 data points, respectively, the semi-supervised learning model consistently outperformed the supervised learning model, achieving an average classification accuracy of 80.9%. The preliminary findings indicate that applying semi-supervised learning to the development of AI models using DUS signals can achieve high generalization accuracy and reduce the effort. This approach may enhance the quality of fetal monitoring.
RESUMO
BACKGROUND: Cognitive assessment plays a pivotal role in the early detection of cognitive impairment, particularly in the prevention and management of cognitive diseases such as Alzheimer's and Lewy body dementia. Large-scale screening relies heavily on cognitive assessment scales as primary tools, with some low sensitivity and others expensive. Despite significant progress in machine learning for cognitive function assessment, its application in this particular screening domain remains underexplored, often requiring labor-intensive expert annotations. AIMS: This paper introduces a semi-supervised learning algorithm based on pseudo-label with putback (SS-PP), aiming to enhance model efficiency in predicting the high risk of cognitive impairment (HR-CI) by utilizing the distribution of unlabeled samples. DATA: The study involved 189 labeled samples and 215,078 unlabeled samples from real world. A semi-supervised classification algorithm was designed and evaluated by comparison with supervised methods composed by 14 traditional machine-learning methods and other advanced semi-supervised algorithms. RESULTS: The optimal SS-PP model, based on GBDT, achieved an AUC of 0.947. Comparative analysis with supervised learning models and semi-supervised methods demonstrated an average AUC improvement of 8% and state-of-art performance, repectively. CONCLUSION: This study pioneers the exploration of utilizing limited labeled data for HR-CI predictions and evaluates the benefits of incorporating physical examination data, holding significant implications for the development of cost-effective strategies in relevant healthcare domains.
RESUMO
BACKGROUND: Risk prediction plays a crucial role in planning for prevention, monitoring, and treatment. Electronic Health Records (EHRs) offer an expansive repository of temporal medical data encompassing both risk factors and outcome indicators essential for effective risk prediction. However, challenges emerge due to the lack of readily available gold-standard outcomes and the complex effects of various risk factors. Compounding these challenges are the false positives in diagnosis codes, and formidable task of pinpointing the onset timing in annotations. OBJECTIVE: We develop a Semi-supervised Double Deep Learning Temporal Risk Prediction (SeDDLeR) algorithm based on extensive unlabeled longitudinal Electronic Health Records (EHR) data augmented by a limited set of gold standard labels on the binary status information indicating whether the clinical event of interest occurred during the follow-up period. METHODS: The SeDDLeR algorithm calculates an individualized risk of developing future clinical events over time using each patient's baseline EHR features via the following steps: (1) construction of an initial EHR-derived surrogate as a proxy for the onset status; (2) deep learning calibration of the surrogate along gold-standard onset status; and (3) semi-supervised deep learning for risk prediction combining calibrated surrogates and gold-standard onset status. To account for missing onset time and heterogeneous follow-up, we introduce temporal kernel weighting. We devise a Gated Recurrent Units (GRUs) module to capture temporal characteristics. We subsequently assess our proposed SeDDLeR method in simulation studies and apply the method to the Massachusetts General Brigham (MGB) Biobank to predict type 2 diabetes (T2D) risk. RESULTS: SeDDLeR outperforms benchmark risk prediction methods, including Semi-parametric Transformation Model (STM) and DeepHit, with consistently best accuracy across experiments. SeDDLeR achieved the best C-statistics ( 0.815, SE 0.023; vs STM +.084, SE 0.030, P-value .004; vs DeepHit +.055, SE 0.027, P-value .024) and best average time-specific AUC (0.778, SE 0.022; vs STM + 0.059, SE 0.039, P-value .067; vs DeepHit + 0.168, SE 0.032, P-value <0.001) in the MGB T2D study. CONCLUSION: SeDDLeR can train robust risk prediction models in both real-world EHR and synthetic datasets with minimal requirements of labeling event times. It holds the potential to be incorporated for future clinical trial recruitment or clinical decision-making.
RESUMO
Wound management requires the measurement of the wound parameters such as its shape and area. However, computerized analysis of the wound suffers the challenge of inexact segmentation of the wound images due to limited or inaccurate labels. It is a common scenario that the source domain provides an abundance of labeled data, while the target domain provides only limited labels. To overcome this, we propose a novel approach that combines self-training learning and mixup augmentation. The neural network is trained on the source domain to generate weak labels on the target domain via the self-training process. In the second stage, generated labels are mixed up with labels from the source domain to retrain the neural network and enhance generalization across diverse datasets. The efficacy of our approach was evaluated using the DFUC 2022, FUSeg, and RMIT datasets, demonstrating substantial improvements in segmentation accuracy and robustness across different data distributions. Specifically, in single-domain experiments, segmentation on the DFUC 2022 dataset scored a dice score of 0.711, while the score on the FUSeg dataset achieved 0.859. For domain adaptation, when these datasets were used as target datasets, the dice scores were 0.714 for DFUC 2022 and 0.561 for FUSeg.
RESUMO
Large labeled data bring significant performance improvement, but acquiring labeled medical data is particularly challenging due to the laborious, time-consuming, and medically qualified annotation. Semi-supervised learning has been employed to leverage unlabeled data. However, the quality and quantity of annotated data have a great influence on the performance of the semi-supervised model. Selecting informative samples through active learning is crucial and could improve model performance. Therefore, we propose a unified semi-supervised active learning architecture (RL-based SSAL) that alternately trains a semi-supervised network and performs active sample selection. Semi-supervised model is first well trained for sample selection, and selected label-required samples are annotated and added to the previously labeled dataset for subsequent semi-supervised model training. To learn to select the most informative samples, we adopt a policy learning-based approach that treats sample selection as a decision-making process. A novel reward function based on the product of predictive confidence and uncertainty is designed, aiming to select samples with high confidence and uncertainty. Comparisons with a semi-supervised baseline on collected lumbar disc herniation dataset demonstrate the effectiveness of the proposed RL-based SSAL, achieving over 3% promotion across different amounts of labeled data. Comparisons with other active learning methods and ablation studies reveal the superiority of proposed policy learning based on active sample selection and reward function. Our model trained with only 200 labeled data achieves an accuracy of 89.32% which is comparable to the performance achieved with the entire labeled dataset, demonstrating its significant advantage.
RESUMO
Pathological images are considered the gold standard for clinical diagnosis and cancer grading. Automatic segmentation of pathological images is a fundamental and crucial step in constructing powerful computer-aided diagnostic systems. Medical microscopic hyperspectral pathological images can provide additional spectral information, further distinguishing different chemical components of biological tissues, offering new insights for accurate segmentation of pathological images. However, hyperspectral pathological images have higher resolution and larger area, and their annotation requires more time and clinical experience. The lack of precise annotations limits the progress of research in pathological image segmentation. In this paper, we propose a novel semi-supervised segmentation method for microscopic hyperspectral pathological images based on multi-consistency learning (MCL-Net), which combines consistency regularization methods with pseudo-labeling techniques. The MCL-Net architecture employs a shared encoder and multiple independent decoders. We introduce a Soft-Hard pseudo-label generation strategy in MCL-Net to generate pseudo-labels that are closer to real labels for pathological images. Furthermore, we propose a multi-consistency learning strategy, treating pseudo-labels generated by the Soft-Hard process as real labels, by promoting consistency between predictions of different decoders, enabling the model to learn more sample features. Extensive experiments in this paper demonstrate the effectiveness of the proposed method, providing new insights for the segmentation of microscopic hyperspectral tissue pathology images.
RESUMO
Inferring cell type proportions from bulk transcriptome data is crucial in immunology and oncology. Here, we introduce guided LDA deconvolution (GLDADec), a bulk deconvolution method that guides topics using cell type-specific marker gene names to estimate topic distributions for each sample. Through benchmarking using blood-derived datasets, we demonstrate its high estimation performance and robustness. Moreover, we apply GLDADec to heterogeneous tissue bulk data and perform comprehensive cell type analysis in a data-driven manner. We show that GLDADec outperforms existing methods in estimation performance and evaluate its biological interpretability by examining enrichment of biological processes for topics. Finally, we apply GLDADec to The Cancer Genome Atlas tumor samples, enabling subtype stratification and survival analysis based on estimated cell type proportions, thus proving its practical utility in clinical settings. This approach, utilizing marker gene names as partial prior information, can be applied to various scenarios for bulk data deconvolution. GLDADec is available as an open-source Python package at https://github.com/mizuno-group/GLDADec.
Assuntos
Software , Humanos , Perfilação da Expressão Gênica/métodos , Algoritmos , Transcriptoma , Biologia Computacional/métodos , Neoplasias/genética , Biomarcadores Tumorais/genética , Marcadores GenéticosRESUMO
The switch machine, an essential element of railway infrastructure, is crucial in maintaining the safety of railway operations. Traditional methods for fault diagnosis are constrained by their dependence on extensive labeled datasets. Semi-supervised learning (SSL), although a promising solution to the scarcity of samples, faces challenges such as the imbalance of pseudo-labels and inadequate data representation. In response, this paper presents the Semi-Supervised Adaptive Matrix Machine (SAMM) model, designed for the fault diagnosis of switch machine. SAMM amalgamates semi-supervised learning with adaptive technologies, leveraging adaptive low-rank regularizer to discern the fundamental links between the rows and columns of matrix data and applying adaptive penalty items to correct imbalances across sample categories. This model methodically enlarges its labeled dataset using probabilistic outputs and semi-supervised, automatically adjusting parameters to accommodate diverse data distributions and structural nuances. The SAMM model's optimization process employs the alternating direction method of multipliers (ADMM) to identify solutions efficiently. Experimental evidence from a dataset containing current signals from switch machines indicates that SAMM outperforms existing baseline models, demonstrating its exceptional status diagnostic capabilities in situations where labeled samples are scarce. Consequently, SAMM offers an innovative and effective approach to semi-supervised classification tasks involving matrix data.
RESUMO
Automated perimetrium segmentation of transvaginal ultrasound images is an important process for computer-aided diagnosis of uterine diseases. However, ultrasound images often contain various structures and textures, and these structures have different shapes, sizes, and contrasts; therefore, accurately segmenting the parametrium region of the uterus in transvaginal uterine ultrasound images is a challenge. Recently, many fully supervised deep learning-based methods have been proposed for the segmentation of transvaginal ultrasound images. Nevertheless, these methods require extensive pixel-level annotation by experienced sonographers. This procedure is expensive and time-consuming. In this paper, we present a bidirectional copy-paste Mamba (BCP-Mamba) semi-supervised model for segmenting the parametrium. The proposed model is based on a bidirectional copy-paste method and incorporates a U-shaped structure model with a visual state space (VSS) module instead of the traditional sampling method. A dataset comprising 1940 transvaginal ultrasound images from Tongji Hospital, Huazhong University of Science and Technology is utilized for training and evaluation. The proposed BCP-Mamba model undergoes comparative analysis with two widely recognized semi-supervised models, BCP-Net and U-Net, across various evaluation metrics including Dice, Jaccard, average surface distance (ASD), and Hausdorff_95. The results indicate the superior performance of the BCP-Mamba semi-supervised model, achieving a Dice coefficient of 86.55%, surpassing both U-Net (80.72%) and BCP-Net (84.63%) models. The Hausdorff_95 of the proposed method is 14.56. In comparison, the counterparts of U-Net and BCP-Net are 23.10 and 21.34, respectively. The experimental findings affirm the efficacy of the proposed semi-supervised learning approach in segmenting transvaginal uterine ultrasound images. The implementation of this model may alleviate the expert workload and facilitate more precise prediction and diagnosis of uterine-related conditions.
RESUMO
This paper presents an on-device semi-supervised human activity detection system that can learn and predict human activity patterns in real time. The clinical objective is to monitor and detect the unhealthy sedentary lifestyle of a user. The proposed semi-supervised learning (SSL) framework uses sparsely labelled user activity events acquired from Inertial Measurement Unit sensors installed as wearable devices. The proposed cluster-based learning model in this approach is trained with data from the same target user, thus preserving data privacy while providing personalized activity detection services. Two different cluster labelling strategies, namely, population-based and distance-based strategies, are employed to achieve the desired classification performance. The proposed system is shown to be highly accurate and computationally efficient for different algorithmic parameters, which is relevant in the context of limited computing resources on typical wearable devices. Extensive experimentation and simulation study have been conducted on multi-user human activity data from the public domain in order to analyze the trade-off between classification accuracy and computation complexity of the proposed learning paradigm with different algorithmic hyper-parameters. With 4.17 h of training time for 8000 activity episodes, the proposed SSL approach consumes at most 20 KB of CPU memory space, while providing a maximum accuracy of 90% and 100% classification rates.