RESUMEN
The transformation model with partly interval-censored data offers a highly flexible modeling framework that can simultaneously support multiple common survival models and a wide variety of censored data types. However, the real data may contain unexplained heterogeneity that cannot be entirely explained by covariates and may be brought on by a variety of unmeasured regional characteristics. Due to this, we introduce the conditionally autoregressive prior into the transformation model with partly interval-censored data and take the spatial frailty into account. An efficient Markov chain Monte Carlo method is proposed to handle the posterior sampling and model inference. The approach is simple to use and does not include any challenging Metropolis steps owing to four-stage data augmentation. Through several simulations, the suggested method's empirical performance is assessed and then the method is used in a leukemia study.
RESUMEN
BACKGROUND: Postoperative hypoparathyroidism is a major complication of thyroidectomy, occurring when the parathyroid glands are inadvertently damaged during surgery. Although intraoperative images are rarely used to train artificial intelligence (AI) because of its complex nature, AI may be trained to intraoperatively detect parathyroid glands using various augmentation methods. The purpose of this study was to train an effective AI model to detect parathyroid glands during thyroidectomy. METHODS: Video clips of the parathyroid gland were collected during thyroid lobectomy procedures. Confirmed parathyroid images were used to train three types of datasets according to augmentation status: baseline, geometric transformation, and generative adversarial network-based image inpainting. The primary outcome was the average precision of the performance of AI in detecting parathyroid glands. RESULTS: 152 Fine-needle aspiration-confirmed parathyroid gland images were acquired from 150 patients who underwent unilateral lobectomy. The average precision of the AI model in detecting parathyroid glands based on baseline data was 77%. This performance was enhanced by applying both geometric transformation and image inpainting augmentation methods, with the geometric transformation data augmentation dataset showing a higher average precision (79%) than the image inpainting model (78.6%). When this model was subjected to external validation using a completely different thyroidectomy approach, the image inpainting method was more effective (46%) than both the geometric transformation (37%) and baseline (33%) methods. CONCLUSION: This AI model was found to be an effective and generalizable tool in the intraoperative identification of parathyroid glands during thyroidectomy, especially when aided by appropriate augmentation methods. Additional studies comparing model performance and surgeon identification, however, are needed to assess the true clinical relevance of this AI model.
RESUMEN
Objective.Cardiovascular diseases are a major cause of mortality globally, and electrocardiograms (ECGs) are crucial for diagnosing them. Traditionally, ECGs are stored in printed formats. However, these printouts, even when scanned, are incompatible with advanced ECG diagnosis software that require time-series data. Digitizing ECG images is vital for training machine learning models in ECG diagnosis, leveraging the extensive global archives collected over decades. Deep learning models for image processing are promising in this regard, although the lack of clinical ECG archives with reference time-series data is challenging. Data augmentation techniques using realistic generative data models provide a solution.Approach.We introduceECG-Image-Kit, an open-source toolbox for generating synthetic multi-lead ECG images with realistic artifacts from time-series data, aimed at automating the conversion of scanned ECG images to ECG data points. The tool synthesizes ECG images from real time-series data, applying distortions like text artifacts, wrinkles, and creases on a standard ECG paper background.Main results.As a case study, we used ECG-Image-Kit to create a dataset of 21 801 ECG images from the PhysioNet QT database. We developed and trained a combination of a traditional computer vision and deep neural network model on this dataset to convert synthetic images into time-series data for evaluation. We assessed digitization quality by calculating the signal-to-noise ratio and compared clinical parameters like QRS width, RR, and QT intervals recovered from this pipeline, with the ground truth extracted from ECG time-series. The results show that this deep learning pipeline accurately digitizes paper ECGs, maintaining clinical parameters, and highlights a generative approach to digitization.Significance.The toolbox has broad applications, including model development for ECG image digitization and classification. The toolbox currently supports data augmentation for the 2024 PhysioNet Challenge, focusing on digitizing and classifying paper ECG images.
Asunto(s)
Aprendizaje Profundo , Electrocardiografía , Procesamiento de Imagen Asistido por Computador , Procesamiento de Imagen Asistido por Computador/métodos , Humanos , Procesamiento de Señales Asistido por Computador , Artefactos , Programas InformáticosRESUMEN
Introduction: Extracting beat-by-beat information from electrocardiograms (ECGs) is crucial for various downstream diagnostic tasks that rely on ECG-based measurements. However, these measurements can be expensive and time-consuming to produce, especially for long-term recordings. Traditional ECG detection and delineation methods, relying on classical signal processing algorithms such as those based on wavelet transforms, produce high-quality delineations but struggle to generalise to diverse ECG patterns. Machine learning (ML) techniques based on deep learning algorithms have emerged as promising alternatives, capable of achieving similar performance without handcrafted features or thresholds. However, supervised ML techniques require large annotated datasets for training, and existing datasets for ECG detection/delineation are limited in size and the range of pathological conditions they represent. Methods: This article addresses this challenge by introducing two key innovations. First, we develop a synthetic data generation scheme that probabilistically constructs unseen ECG traces from "pools" of fundamental segments extracted from existing databases. A set of rules guides the arrangement of these segments into coherent synthetic traces, while expert domain knowledge ensures the realism of the generated traces, increasing the input variability for training the model. Second, we propose two novel segmentation-based loss functions that encourage the accurate prediction of the number of independent ECG structures and promote tighter segmentation boundaries by focusing on a reduced number of samples. Results: The proposed approach achieves remarkable performance, with a F 1 -score of 99.38% and delineation errors of 2.19 ± 17.73 ms and 4.45 ± 18.32 ms for ECG segment onsets and offsets across the P, QRS, and T waves. These results, aggregated from three diverse freely available databases (QT, LU, and Zhejiang), surpass current state-of-the-art detection and delineation approaches. Discussion: Notably, the model demonstrated exceptional performance despite variations in lead configurations, sampling frequencies, and represented pathophysiology mechanisms, underscoring its robust generalisation capabilities. Real-world examples, featuring clinical data with various pathologies, illustrate the potential of our approach to streamline ECG analysis across different medical settings, fostered by releasing the codes as open source.
RESUMEN
Defecation care for disabled patients is a major challenge in health management. Traditional post-defecation treatment will bring physical injury and negative emotions to patients, while existing pre-defecation forecasting care methods are physically intrusive. On the basis of exploring the mechanism of defecation intention generation, and based on the characteristic analysis and clinical application of bowel sounds, it is found that the generation of desire to defecate and bowel sounds are correlated to a certain extent. Therefore, a deep learning-based bowel sound recognition method is proposed for human defecation prediction. The wavelet domain based Wiener filter is used to filter the bowel sound data to reduce other noise. Statistical analysis, fast Fourier transform and wavelet packet transform are used to extract the integrated features of bowel sound in time, frequency and time-frequency domain. In particular, an audio signal expansion data algorithm based on the Informer model is proposed to solve the problem of poor generalization of the training model caused by the difficulty of collecting bowel sound in reality. An improved one-dimensional residual network model (1D-IResNet) for defecation classification prediction is designed based on multi-domain features. The experimental results show that the proposed bowel sound augmentation strategy can effectively improve the data sample size and increase the sample diversity. Under the augmented dataset, the training speed of the 1D-IResNet model is accelerated, and the classification accuracy reaches 90.54 %, the F1 score reaches 83.88 %, which achieves a relatively good classification stability while maintaining a high classification index.
RESUMEN
Sequential recommendation typically utilizes deep neural networks to mine rich information in interaction sequences. However, existing methods often face the issue of insufficient interaction data. To alleviate the sparsity issue, self-supervised learning is introduced into sequential recommendation. Despite its effectiveness, we argue that current self-supervised learning-based (i.e., SSL-based) sequential recommendation models have the following limitations: (1) using only a single self-supervised learning method, either contrastive self-supervised learning or generative self-supervised learning. (2) employing a simple data augmentation strategy in either the graph structure domain or the node feature domain. We believe that they have not fully utilized the capabilities of both self-supervised methods and have not sufficiently explored the advantages of combining graph augmentation schemes. As a result, they often fail to learn better item representations. In light of this, we propose a novel multi-task sequential recommendation framework named Adaptive Self-supervised Learning for sequential Recommendation (ASLRec). Specifically, our framework combines contrastive and generative self-supervised learning methods adaptively, simultaneously applying different perturbations at both the graph topology and node feature levels. This approach constructs diverse augmented graph views and employs multiple loss functions (including contrastive loss, generative loss, mask loss, and prediction loss) for joint training. By encompassing the capabilities of various methods, our model learns item representations across different augmented graph views to achieve better performance and effectively mitigate interaction noise and sparsity. In addition, we add a small proportion of random uniform noise to item representations, making the item representations more uniform and mitigating the inherent popularity bias in interaction records. We conduct extensive experiments on three publicly available benchmark datasets to evaluate our model. The results demonstrate that our approach achieves state-of-the-art performance compared to 14 other competitive methods: the hit rate (HR) improved by over 14.39%, and the normalized discounted cumulative gain (NDCG) increased by over 18.67%.
RESUMEN
BACKGROUND: Renal cell carcinoma (RCC) remains a global health concern due to its poor survival rate. This study aimed to investigate the influence of medical determinants and socioeconomic status on survival outcomes of RCC patients. We analyzed the survival data of 41,563 RCC patients recorded under the Surveillance, Epidemiology, and End Results (SEER) program from 2012 to 2020. METHODS: We employed a competing risk model, assuming lifetime of RCC patients under various risks follows Chen distribution. This model accounts for uncertainty related to survival time as well as causes of death, including missing cause of death. For model analysis, we utilized Bayesian inference and obtained the estimate of various key parameters such as cumulative incidence function (CIF) and cause-specific hazard. Additionally, we performed Bayesian hypothesis testing to assess the impact of multiple factors on the survival time of RCC patients. RESULTS: Our findings revealed that the survival time of RCC patients is significantly influenced by gender, income, marital status, chemotherapy, tumor size, and laterality. However, we observed no significant effect of race and origin on patient's survival time. The CIF plots indicated a number of important distinctions in incidence of causes of death corresponding to factors income, marital status, race, chemotherapy, and tumor size. CONCLUSIONS: The study highlights the impact of various medical and socioeconomic factors on survival time of RCC patients. Moreover, it also demonstrates the utility of competing risk model for survival analysis of RCC patients under Bayesian paradigm. This model provides a robust and flexible framework to deal with missing data, which can be particularly useful in real-life situations where patients information might be incomplete.
RESUMEN
Sleep is an essential part of human life, and the quality of one's sleep is also an important indicator of one's health. Analyzing the Electroencephalogram (EEG) signals of a person during sleep makes it possible to understand the sleep status and give relevant rest or medical advice. In this paper, a decent amount of artificial data generated with a data augmentation method based on Discrete Cosine Transform from a small amount of real experimental data of a specific individual is introduced. A classification model with an accuracy of 92.85% has been obtained. By mixing the data augmentation with the public database and training with the EEGNet, we obtained a classification model with significantly higher accuracy for the specific individual. The experiments have demonstrated that we can circumvent the subject-independent problem in sleep EEG in this way and use only a small amount of labeled data to customize a dedicated classification model with high accuracy.
RESUMEN
Skin cancer stands as one of the foremost challenges in oncology, with its early detection being crucial for successful treatment outcomes. Traditional diagnostic methods depend on dermatologist expertise, creating a need for more reliable, automated tools. This study explores deep learning, particularly Convolutional Neural Networks (CNNs), to enhance the accuracy and efficiency of skin cancer diagnosis. Leveraging the HAM10000 dataset, a comprehensive collection of dermatoscopic images encompassing a diverse range of skin lesions, this study introduces a sophisticated CNN model tailored for the nuanced task of skin lesion classification. The model's architecture is intricately designed with multiple convolutional, pooling, and dense layers, aimed at capturing the complex visual features of skin lesions. To address the challenge of class imbalance within the dataset, an innovative data augmentation strategy is employed, ensuring a balanced representation of each lesion category during training. Furthermore, this study introduces a CNN model with optimized layer configuration and data augmentation, significantly boosting diagnostic precision in skin cancer detection. The model's learning process is optimized using the Adam optimizer, with parameters fine-tuned over 50 epochs and a batch size of 128 to enhance the model's ability to discern subtle patterns in the image data. A Model Checkpoint callback ensures the preservation of the best model iteration for future use. The proposed model demonstrates an accuracy of 97.78% with a notable precision of 97.9%, recall of 97.9%, and an F2 score of 97.8%, underscoring its potential as a robust tool in the early detection and classification of skin cancer, thereby supporting clinical decision-making and contributing to improved patient outcomes in dermatology.
Asunto(s)
Aprendizaje Profundo , Dermoscopía , Redes Neurales de la Computación , Neoplasias Cutáneas , Humanos , Neoplasias Cutáneas/diagnóstico por imagen , Neoplasias Cutáneas/patología , Dermoscopía/métodos , Interpretación de Imagen Asistida por Computador/métodosRESUMEN
Performing data augmentation in medical named entity recognition (NER) is crucial due to the unique challenges posed by this field. Medical data is characterized by high acquisition costs, specialized terminology, imbalanced distributions, and limited training resources. These factors make achieving high performance in medical NER particularly difficult. Data augmentation methods help to mitigate these issues by generating additional training samples, thus balancing data distribution, enriching the training dataset, and improving model generalization. This paper proposes two data augmentation methods-Contextual Random Replacement based on Word2Vec Augmentation (CRR) and Targeted Entity Random Replacement Augmentation (TER)-aimed at addressing the scarcity and imbalance of data in the medical domain. When combined with a deep learning-based Chinese NER model, these methods can significantly enhance performance and recognition accuracy under limited resources. Experimental results demonstrate that both augmentation methods effectively improve the recognition capability of medical named entities. Specifically, the BERT-BiLSTM-CRF model achieved the highest F1 score of 83.587%, representing a 1.49% increase over the baseline model. This validates the importance and effectiveness of data augmentation in medical NER.
Asunto(s)
Aprendizaje Profundo , Humanos , Procesamiento de Lenguaje NaturalRESUMEN
Different pathologies of the hip are characterized by the abnormal shape of the bony structures of the joint, namely the femur and the acetabulum. Three-dimensional (3D) models of the hip can be used for diagnosis, biomechanical simulation, and planning of surgical treatments. These models can be generated by building 3D surfaces of the joint's structures segmented on magnetic resonance (MR) images. Deep learning can avoid time-consuming manual segmentations, but its performance depends on the amount and quality of the available training data. Data augmentation and transfer learning are two approaches used when there is only a limited number of datasets. In particular, data augmentation can be used to artificially increase the size and diversity of the training datasets, whereas transfer learning can be used to build the desired model on top of a model previously trained with similar data. This study investigates the effect of data augmentation and transfer learning on the performance of deep learning for the automatic segmentation of the femur and acetabulum on 3D MR images of patients diagnosed with femoroacetabular impingement. Transfer learning was applied starting from a model trained for the segmentation of the bony structures of the shoulder joint, which bears some resemblance to the hip joint. Our results suggest that data augmentation is more effective than transfer learning, yielding a Dice similarity coefficient compared to ground-truth manual segmentations of 0.84 and 0.89 for the acetabulum and femur, respectively, whereas the Dice coefficient was 0.78 and 0.88 for the model based on transfer learning. The Accuracy for the two anatomical regions was 0.95 and 0.97 when using data augmentation, and 0.87 and 0.96 when using transfer learning. Data augmentation can improve the performance of deep learning models by increasing the diversity of the training dataset and making the models more robust to noise and variations in image quality. The proposed segmentation model could be combined with radiomic analysis for the automatic evaluation of hip pathologies.
RESUMEN
Accurate segmentation of chronic stroke lesions from mono-spectral magnetic resonance imaging scans (e.g., T1-weighted images) is a difficult task due to the arbitrary shape, complex texture, variable size and intensities, and varied locations of the lesions. Due to this inherent spatial heterogeneity, existing machine learning methods have shown moderate performance for chronic lesion delineation. In this study, we introduced: (1) a method that integrates transformers' deformable feature attention mechanism with convolutional deep learning architecture to improve the accuracy and generalizability of stroke lesion segmentation, and (2) an ecological data augmentation technique based on inserting real lesions into intact brain regions. Our combination of these two approaches resulted in a significant increase in segmentation performance, with a Dice index of 0.82 (±0.39), outperforming the existing methods trained and tested on the same Anatomical Tracings of Lesions After Stroke (ATLAS) 2022 dataset. Our method performed relatively well even for cases with small stroke lesions. We validated the robustness of our method through an ablation study and by testing it on new unseen brain scans from the Ischemic Stroke Lesion Segmentation (ISLES) 2015 dataset. Overall, our proposed approach of transformers with ecological data augmentation offers a robust way to delineate chronic stroke lesions with clinically relevant accuracy. Our method can be extended to other challenging tasks that require automated detection and segmentation of diverse brain abnormalities from clinical scans.
Asunto(s)
Aprendizaje Profundo , Imagen por Resonancia Magnética , Accidente Cerebrovascular , Humanos , Imagen por Resonancia Magnética/métodos , Imagen por Resonancia Magnética/normas , Accidente Cerebrovascular/diagnóstico por imagen , Accidente Cerebrovascular/patología , Neuroimagen/métodos , Neuroimagen/normas , Accidente Cerebrovascular Isquémico/diagnóstico por imagen , Procesamiento de Imagen Asistido por Computador/métodos , Anciano , Encéfalo/diagnóstico por imagen , Encéfalo/patologíaRESUMEN
Large-scale, diverse, and high-quality data are the basis and key to achieving a good generalization of target detection and recognition algorithms based on deep learning. However, the existing methods for the intelligent augmentation of synthetic aperture radar (SAR) images are confronted with several issues, including training instability, inferior image quality, lack of physical interpretability, etc. To solve the above problems, this paper proposes a feature-level SAR target-data augmentation method. First, an enhanced capsule neural network (CapsNet) is proposed and employed for feature extraction, decoupling the attribute information of input data. Moreover, an attention mechanism-based attribute decoupling framework is used, which is beneficial for achieving a more effective representation of features. After that, the decoupled attribute feature, including amplitude, elevation angle, azimuth angle, and shape, can be perturbed to increase the diversity of features. On this basis, the augmentation of SAR target images is realized by reconstructing the perturbed features. In contrast to the augmentation methods using random noise as input, the proposed method realizes the mapping from the input of known distribution to the change in unknown distribution. This mapping method reduces the correlation distance between the input signal and the augmented data, therefore diminishing the demand for training data. In addition, we combine pixel loss and perceptual loss in the reconstruction process, which improves the quality of the augmented SAR data. The evaluation of the real and augmented images is conducted using four assessment metrics. The images generated by this method achieve a peak signal-to-noise ratio (PSNR) of 21.6845, radiometric resolution (RL) of 3.7114, and dynamic range (DR) of 24.0654. The experimental results demonstrate the superior performance of the proposed method.
RESUMEN
In recent years, deep learning (DL) techniques have achieved remarkable success in various fields of computer vision. This progress was attributed to the vast amounts of data utilized to train these models, as they facilitated the learning of more intricate and detailed feature information about target objects, leading to improved model performance. However, in most real-world tasks, it was challenging to gather sufficient data for model training. Insufficient datasets often resulted in models prone to overfitting. To address this issue and enhance model performance, generalization ability, and mitigate overfitting in data-limited scenarios, image data augmentation methods have been proposed. These methods generated synthetic samples to augment the original dataset, emerging as a preferred strategy to boost model performance when data was scarce. This review first introduced commonly used and highly effective image data augmentation techniques, along with a detailed analysis of their advantages and disadvantages. Second, this review presented several datasets frequently employed for evaluating the performance of image data augmentation methods and examined how advanced augmentation techniques can enhance model performance. Third, this review discussed the applications and performance of data augmentation techniques in various computer vision domains. Finally, this review provided an outlook on potential future research directions for image data augmentation methods.
RESUMEN
Plant leaf diseases are a significant concern in agriculture due to their detrimental impact on crop productivity and food security. Effective disease management depends on the early and accurate detection and diagnosis of these conditions, facilitating timely intervention and mitigation strategies. In this study, we address the pressing need for accurate and efficient methods for detecting leaf diseases by introducing a new architecture called DenseNet201Plus. DenseNet201 was modified by including superior data augmentation and pre-processing techniques, an attention-based transition mechanism, multiple attention modules, and dense blocks. These modifications enhance the robustness and accuracy of the proposed DenseNet201Plus model in diagnosing diseases related to plant leaves. The proposed architecture was trained using two distinct datasets: Banana Leaf Disease and Black Gram Leaf Disease. Through extensive experimentation, we evaluated the performance of DenseNet201Plus in terms of various classification metrics and achieved values of 0.9012, 0.9012, 0.9012, and 0.9716 for accuracy, precision, recall, and AUC for the banana leaf disease dataset, respectively. Similarly, the black gram leaf disease dataset model provides values of 0.9950, 0.9950, 0.9950, and 1.0 for accuracy, precision, recall, and AUC. Compared to other well-known pre-trained convolutional neural network (CNN) architectures, our proposed model demonstrates superior performance in both utilized datasets. Last but not least, we combined the strength of Grad-CAM++ with our proposed model to enhance the interpretability and localization of disease areas, providing valuable insights for agricultural practitioners and researchers to make informed decisions and optimize disease management strategies.
RESUMEN
Segmentation in medical images is inherently ambiguous. It is crucial to capture the uncertainty in lesion segmentations to assist cancer diagnosis and further interventions. Recent works have made great progress in generating multiple plausible segmentation results as diversified references to account for the uncertainty in lesion segmentations. However, the efficiency of existing models is limited, and the uncertainty information lying in multi-annotated datasets remains to be fully utilized. In this study, we propose a series of methods to corporately deal with the above limitation and leverage the abundant information in multi-annotated datasets: (1) Customized T-time Inner Sampling Network to promote the modeling flexibility and efficiently generate samples matching the ground-truth distribution of a number of annotators; (2) Uncertainty Degree defined for quantitatively measuring the uncertainty of each sample and the imbalance of the whole multi-annotated dataset from a brand new perspective; (3) Uncertainty-aware Data Augmentation Strategy to help probabilistic models adaptively fit samples with different ranges of uncertainty. We have evaluated each of them on both the publicly available lung nodule dataset and our in-house Liver Tumor dataset. Results show that our proposed methods achieves the overall best performance on both accuracy and efficiency, demonstrating its great potential in lesion segmentations and more downstream tasks in real clinical scenarios.
Asunto(s)
Neoplasias Pulmonares , Humanos , Incertidumbre , Neoplasias Pulmonares/diagnóstico por imagen , Tomografía Computarizada por Rayos X/métodos , Algoritmos , Bases de Datos FactualesRESUMEN
Gait analysis has proven to be a key process in the functional assessment of people involving many fields, such as diagnosis of diseases or rehabilitation, and has increased in relevance lately. Gait analysis often requires gathering data, although this can be very expensive and time consuming. One of the main solutions applied in fields when data acquisition is a problem is augmentation of datasets with artificial data. There are two main approaches for doing that: simulation and synthetic data generation. In this article, we propose a parametrizable generative system of synthetic walking simplified human skeletons. For achieving that, a data gathering experiment with up to 26 individuals was conducted. The system consists of two artificial neural networks: a recurrent neural network for the generation of the movement and a multilayer perceptron for determining the size of the segments of the skeletons. The system has been evaluated through four processes: (i) an observational appraisal by researchers in gait analysis, (ii) a visual representation of the distribution of the generated data, (iii) a numerical analysis using the normalized cross-correlation coefficient, and (iv) an angular evaluation to check the kinematic validity of the data. The evaluation concluded that the system is able to generate realistic and accurate gait data. These results reveal a promising path for this research field, which can be further improved through increasing the variety of movements and the user sample.
Asunto(s)
Redes Neurales de la Computación , Humanos , Marcha/fisiología , Modelos Biológicos , Fenómenos Biomecánicos/fisiología , Masculino , Caminata/fisiología , FemeninoRESUMEN
Objective.Due to the difficulty in acquiring motor imagery electroencephalography (MI-EEG) data and ensuring its quality, insufficient training data often leads to overfitting and inadequate generalization capabilities of deep learning-based classification networks. Therefore, we propose a novel data augmentation method and deep learning classification model to enhance the decoding performance of MI-EEG further.Approach.The raw EEG signals were transformed into the time-frequency maps as the input to the model by continuous wavelet transform. An improved Wasserstein generative adversarial network with gradient penalty data augmentation method was proposed, effectively expanding the dataset used for model training. Additionally, a concise and efficient deep learning model was designed to improve decoding performance further.Main results.It has been demonstrated through validation by multiple data evaluation methods that the proposed generative network can generate more realistic data. Experimental results on the BCI Competition IV 2a and 2b datasets and the actual collected dataset show that classification accuracies are 83.4%, 89.1% and 73.3%, and Kappa values are 0.779, 0.782 and 0.644, respectively. The results indicate that the proposed model outperforms state-of-the-art methods.Significance.Experimental results demonstrate that this method effectively enhances MI-EEG data, mitigates overfitting in classification networks, improves MI classification accuracy, and holds positive implications for MI tasks.
Asunto(s)
Interfaces Cerebro-Computador , Electroencefalografía , Imaginación , Redes Neurales de la Computación , Electroencefalografía/métodos , Electroencefalografía/clasificación , Humanos , Imaginación/fisiología , Aprendizaje Profundo , Análisis de OndículasRESUMEN
Deep learning has emerged as a robust tool for automating feature extraction from three-dimensional images, offering an efficient alternative to labour-intensive and potentially biased manual image segmentation methods. However, there has been limited exploration into the optimal training set sizes, including assessing whether artficial expansion by data augmentation can achieve consistent results in less time and how consistent these benefits are across different types of traits. In this study, we manually segmented 50 planktonic foraminifera specimens from the genus Menardella to determine the minimum number of training images required to produce accurate volumetric and shape data from internal and external structures. The results reveal unsurprisingly that deep learning models improve with a larger number of training images with eight specimens being required to achieve 95% accuracy. Furthermore, data augmentation can enhance network accuracy by up to 8.0%. Notably, predicting both volumetric and shape measurements for the internal structure poses a greater challenge compared with the external structure, owing to low contrast differences between different materials and increased geometric complexity. These results provide novel insight into optimal training set sizes for precise image segmentation of diverse traits and highlight the potential of data augmentation for enhancing multivariate feature extraction from three-dimensional images.
RESUMEN
In recent years, hyperspectral imaging combined with machine learning techniques has garnered significant attention for its potential in assessing fruit maturity. This study proposes a method for predicting strawberry fruit maturity based on the harvest time. The main features of this study are as follows. 1) Selection of wavelength band associated with strawberry growth season; 2) Extraction of efficient parameters to predict strawberry maturity 3) Prediction of internal quality attributes of strawberries using extracted parameters. In this study, experts cultivated strawberries in a controlled environment and performed hyperspectral measurements and organic analyses on the fruit with minimal time delay to facilitate accurate modeling. Data augmentation techniques through cross-validation and interpolation were effective in improving model performance. The four parameters included in the model and the cumulative value of the model were available for quality prediction as additional parameters. Among these five parameter candidates, two parameters with linearity were finally identified. The predictive outcomes for firmness, soluble solids content, acidity, and anthocyanin levels in strawberry fruit, based on the two identified parameters, are as follows: The first parameter, ps, demonstrated RMSE performances of 1.0 N, 2.3 %, 0.1 %, and 2.0 mg per 100 g fresh fruit for firmness, soluble solids content, acidity, and anthocyanin, respectively. The second parameter, p3, showed RMSE performances of 0.6 N, 1.2 %, 0.1 %, and 1.8 mg per 100 g fresh fruit, respectively. The proposed non-destructive analysis method shows the potential to overcome the challenges associated with destructive testing methods for assessing certain internal qualities of strawberry fruit.