Fine-tuned Transfer Research Articles

Explainable deep learning identifies patterns and drivers of freshwater harmful algal blooms.

The escalating magnitude, frequency, and duration of harmful algal blooms (HABs) pose significant challenges to freshwater ecosystems worldwide. However, the mechanisms driving HABs remain poorly understood, in part due to the strong regional specificity of algal processes and the uneven data availability. These complexities make it difficult to generalize HAB dynamics and effectively predict their occurrence using traditional models. To address these challenges, we developed an explainable deep learning approach using long short-term memory (LSTM) models combined with explanation techniques that can capture complex patterns and provide explainable insights into key HAB drivers. We applied this approach for algal density modeling at 102 sites in China's lakes and reservoirs over three years. LSTMs effectively captured daily algal dynamics, achieving mean and maximum Nash-Sutcliffe efficiency coefficients of 0.48 and 0.95 during testing phase. Moreover, water temperature emerged as the primary driver of HABs both nationally and in over 30% of localities, with stronger water temperature sensitivity observed in mid-to low-latitudes. We also identified regional similarities that allow for the successful transferability in modeling algal dynamics. Specifically, using fine-tuned transfer learning, we improved the prediction accuracy in over 75% of poorly gauged areas. Overall, LSTM-based explainable deep learning approach effectively addresses key challenges in HAB modeling by tackling both regional specificity and data limitations. By accurately predicting algal dynamics and identifying critical drivers, this approach provides actionable insights into the mechanisms of HABs, ultimately aids in the implementation of effective mitigation measures for nationwide and regional freshwater ecosystems.

Use of Data Augmentation and Fine-tuned Transfer Learning Models for Classification of Cervical Cancer

Innovative strategies for early and accurate diagnosis of cervical cancer continue to be a global health priority. In this research, we present a powerful framework for improving cervical cancer classification by combining data augmentation methods with tailored transfer learning models. The lack of properly labelled medical data is a major roadblock to developing reliable diagnosis models. The geometric transformations, contrast tweaks, and noise addition that we use to increase the dataset artificially help us overcome this difficulty. The model's robustness is improved as a result of the additional exposure to diverse real-world circumstances provided by this supplemented dataset. We use a transfer learning strategy based on pre-trained convolutional neural networks (CNNs) to harness the potential of deep learning. These networks are well-suited for medical image analysis due to their ability to quickly and accurately identify features in a wide variety of images. Our expanded cervical cancer dataset is used to fine-tune these algorithms for their diagnostic purpose. Our experiments show that utilising both data augmentation and fine-tuned transfer learning considerably raises the accuracy of cervical cancer categorization. In terms of sensitivity, specificity, and overall accuracy, the model performs at a state-of-the-art level. This suggests that it may be useful in the diagnosis and early identification of cervical cancer. Additionally, our method demonstrates the significance of efficient data utilisation and model adaptability in the field of medicine. Our method is scalable and affordable, and it can be easily implemented in healthcare settings since we handle the problem of data scarcity and take advantage of the features of pre-trained deep learning models.

Improved Prototypical Network Model for Classification of Farmland Shelterbelt Using Sentinel-2 Imagery

Farmland shelterbelt plays an important role in protecting farmland and ensuring stable crop yields, and it is mainly distributed in the form of bands and patches; different forms of distribution have different impacts on farmland, which is an important factor affecting crop yields. Therefore, high-precision classification of banded and patch farmland shelterbelt is a prerequisite for analyzing its impact on crop yield. In this study, we explored the effectiveness and transferability of an improved Prototypical Network model incorporating data augmentation and a convolutional block attention module for extracting banded and patch farmland shelterbelt in Northeast China, and we analyzed the potential of applying it to the production of large-scale farmland shelterbelt products. Firstly, we classified banded and patch farmland shelterbelt under different sample window sizes using the improved Prototypical Network in the source domain study area to obtain the optimal sample window size and the optimal classification model. Secondly, fine-tuning transfer learning and learning from scratch directly were used to classify the banded and patch farmland shelterbelt in the target domain study area, respectively, to evaluate the extraction model’s migratability. The results showed that classification of farmland shelterbelt using the improved Prototypical Network is very effective, with the highest extraction accuracy under the 5 × 5 sample window; the accuracies of the banded and patch farmland shelterbelt are 92.16% and 90.91%, respectively. Using the fine-tuning transfer learning method in the target domain can classify the banded and patch farmland shelterbelt with high accuracy, above 95% and 89%, respectively. The proposed approach can provide new insight into farmland shelterbelt classification and farmland shelterbelt products obtained from freely accessible Sentinel-2 multispectral images.

Comparative analysis of vision transformers and fine-tuned transfer learning models for brain tumor classification

Brain tumors are a primary factor causing cancer-related deaths globally, and their classification remains a significant research challenge due to the variability in tumor intensity, size, and shape, as well as the similar appearances of different tumor types. Accurate differentiation is further complicated by these factors, making diagnosis difficult even with advanced imaging techniques such as magnetic resonance imaging (MRI). Recent techniques in artificial intelligence (AI), in particular deep learning (DL), have improved the speed and accuracy of medical image analysis, but they still face challenges like overfitting and the need for large annotated datasets. This study addresses these challenges by presenting two approaches for brain tumor classification using MRI images. The first approach involves fine-tuning transfer learning cutting-edge models, including SEResNet, ConvNeXtBase, and ResNet101V2, with global average pooling 2D and dropout layers to minimize overfitting and reduce the need for extensive preprocessing. The second approach leverages the Vision Transformer (ViT), optimized with the AdamW optimizer and extensive data augmentation. Experiments on the BT-Large-4C dataset demonstrate that SEResNet achieves the highest accuracy of 97.96%, surpassing ViT’s 95.4%. These results suggest that fine-tuning and transfer learning models are more effective at addressing the challenges of overfitting and dataset limitations, ultimately outperforming the Vision Transformer and existing state-of-the-art techniques in brain tumor classification.

Skin Cancer Classification Using Fine-Tuned Transfer Learning of DENSENET-121

Skin cancer diagnosis greatly benefits from advanced machine learning techniques, particularly fine-tuned deep learning models. In our research, we explored the impact of traditional machine learning and fine-tuned deep learning approaches on prediction accuracy. Our findings reveal significant improvements in predictability and accuracy with fine-tuning, particularly evident in deep learning models. The CNN, SVM, and Random Forest Classifier achieved high accuracy. However, fine-tuned deep learning models such as EfficientNetB0, ResNet34, VGG16, Inception _v3, and DenseNet121 demonstrated superior performance. To ensure comparability, we fine-tuned these models by incorporating additional layers, including one flatten layer and three densely interconnected layers. These layers play a crucial role in enhancing model efficiency and performance. The flatten layer preprocesses multidimensional feature maps, facilitating efficient information flow, while subsequent dense layers refine feature representations, capturing intricate patterns and relationships within the data. Leveraging LeakyReLU activation functions in the dense layers mitigates the vanishing gradient problem and promotes stable training. Finally, the output dense layer with a sigmoid activation function simplifies decision making for healthcare professionals by providing binary classification output. Our study underscores the significance of incorporating additional layers in fine-tuned neural network models for skin cancer classification, offering improved accuracy and reliability in diagnosis.

Exploring the Trade-Off between generalist and specialized Models: A center-based comparative analysis for glioblastoma segmentation

IntroductionInherent variations between inter-center data can undermine the robustness of segmentation models when applied at a specific center (dataset shift). We investigated whether specialized center-specific models are more effective compared to generalist models based on multi-center data, and how center-specific data could enhance the performance of generalist models within a particular center using a fine-tuning transfer learning approach. For this purpose, we studied the dataset shift at center level and conducted a comparative analysis to assess the impact of data source on glioblastoma segmentation models. Methods & MaterialsThe three key components of dataset shift were studied: prior probability shift—variations in tumor size or tissue distribution among centers; covariate shift—inter-center MRI alterations; and concept shift—different criteria for tumor segmentation. BraTS 2021 dataset was used, which includes 1251 cases from 23 centers. Thereafter, 155 deep-learning models were developed and compared, including 1) generalist models trained with multi-center data, 2) specialized models using only center-specific data, and 3) fine-tuned generalist models using center-specific data. ResultsThe three key components of dataset shift were characterized. The amount of covariate shift was substantial, indicating large variations in MR imaging between different centers. Glioblastoma segmentation models tend to perform best when using data from the application center. Generalist models, trained with over 700 samples, achieved a median Dice score of 88.98%. Specialized models surpassed this with 200 cases, while fine-tuned models outperformed with 50 cases. ConclusionsThe influence of dataset shift on model performance is evident. Fine-tuned and specialized models, utilizing data from the evaluated center, outperform generalist models, which rely on data from other centers. These approaches could encourage medical centers to develop customized models for their local use, enhancing the accuracy and reliability of glioblastoma segmentation in a context where dataset shift is inevitable.

Compression of EEG signals with the LSTM-autoencoder via domain adaptation approach

The successful implementation of neural network-based EEG signal compression has led to significant cost reductions in data transmission. However, a major obstacle in this process arises from the decline in performance when compressing EEG signals from multiple subjects. This challenge arises due to the notable feature shift of EEG signals between subjects, which poses an impediment to the neural network’s efficient concurrent acquisition of information from multiple subjects. To address this limitation and enable more effective utilization of data for improving the performance on target domain, we propose a Domain Adaptation (DA) framework based on LSTM-autoencoder. Our experiments encompassed the following: (1) A comparison between LSTM-autoencoder, GRU-autoencoder, and the commonly used convolutional autoencoder (CAE) in EEG compression. (2) A comparison between our proposed DA method and the MMD-based DA method, as well as Fine-tuning transfer learning. The results demonstrate the following: (1) LSTM-autoencoder outperforms other models in both subject-specific and cross-subject scenarios. (2) Using transfer learning improves the performance of LSTM-autoencoder on the target subject. (3) Our proposed method outperforms maximum mean discrepancy (MMD)-based domain adaptation and fine-tuning approaches, resulting in a more significant enhancement.

Gearbox Fault Diagnosis Based on ICEEMDAN-MPE-AWT and SE-ResNeXt50 Transfer Learning Model

Because the gearbox in transmission systems is prone to failure and the fault signal is not obvious, the fault end cannot be located. In this paper, a gearbox fault diagnosis method grounded on improved complete ensemble empirical mode decomposition with adaptive noise, a multiscale permutation entropy and adaptive wavelet thresholding (ICEEMDAN-MPE-AWT) denoising method and an SE-ResNeXt50 transfer learning model are proposed. Initially, the vibration signal is denoised by ICEEMDAN-MPE-AWT, the denoised vibration signal is then converted into a Gram angle field (GAF) diagram, and then the parameters are transferred by the fine-tuning transfer learning strategy. Finally, a GAF diagram is input into the model for training to achieve fault extraction and classification. In this paper, the open gear dataset of Southeast University is used for experimental research. The experimental results show that when using the ICEEMDAN-MPE-AWT and when the signal-to-noise ratio (SNR) of the experimental data is −4 dB, the average accuracy of the GASF+TSE-ResNeXt50 and the GASF+TSE-ResNeXt18 can reach 98.8% and 97.5%, respectively. When the SNR is 6 dB, the accuracy of the above two models reaches 100% and 99.3%, respectively. Moreover, when compared to alternative approaches, the noise reduction method in this paper can better remove noise interference so that the model can better extract fault features. Therefore, the method proposed in this article shows significant improvement in noise reduction and fault classification accuracy compared to other methods.

Multigrade brain tumor classification in MRI images using Fine tuned efficientnet

Medical imaging plays a vital role in detecting and treating brain tumors. Malignant or non-malignant brain tissue’s abnormal growth causes long-term brain damage. It is crucial to detect and properly categorize the kind of brain tumor. Specialists normally use MRI to create high-contrast grayscale brain images to segment them. Convolutional neural networks (CNN) driven by deep learning (DL) have transformed computer-assisted testing systems by producing good results in a wide range of medical imaging analytics applications, including tumor diagnosis in the brain. The paper introduces a lightweight fine-tuned Convolutional Neural Network EfficientNet ’ECNN’ to detect brain tumors. In this study, we provide a transfer learning-based measurement strategy for grouping cerebrum growths in three distinct datasets with different classifications, such as meningioma, glioma, and pituitary growth, using fine-tuned EfficientNets. The findings of this research rely on Efficient Nets to classify brain tumors in three different types of datasets utilizing a fine-tuned transfer learning mechanism. With EfficientNetV2S as the system’s foundation, our proposed way of fine-tunned pre-trained EfficientNetV2S model outperformed for all datasets over state of the art methods. The effectiveness of the suggested model has been assessed using performance metrics, and outcomes were compared to those produced using state-of-the-art approaches. The average test accuracy, recall, precision, and sensitivity score are 98.48%, 98%, 98.5%, and 98.71%, respectively.

Long-term streamflow forecasting in data-scarce regions: Insightful investigation for leveraging satellite-derived data, Informer architecture, and concurrent fine-tuning transfer learning

Accurate multistep forecasting of the long-term streamflow (Qflow) in poorly gauged basins is pivotal for sustainable water resource management and decision-making. The purpose of this study is to improve the performance of Qflow predictions in data-scarce regions. To achieve this, we employed geo-spatiotemporal satellite-derived datasets, advanced Informer architecture, and transfer learning methodologies. The deep-transfer-learning-based Informer model enables the perception of time dependencies and is specifically tailored to facilitate multistep-ahead Qflow prediction. Initial investigations identified correlations between historical Qflow records and geo-spatiotemporal satellite metrics, and this helped select grids that encapsulated the highest correlation between Qflow and climatic parameters. Subsequent assessments contrasted three main dimensionality reduction techniques—principal component analysis (PCA), t-distributed Stochastic Neighbor Embedding (t-SNE), and multi-dimensional scaling (MDS)—to craft deep-learning-amenable endogenous inputs. The introduction of the current fine-tuning transfer learning (CFTL) methodology, augmented with incremental learning, which was evaluated against conventional transfer learning (CTL) across two distinct target domains with varying similarities to the source domain, was central to this study. A holistic comparison involving a range of univariate multistep-ahead prediction models including DNN, CNN, LSTM, AR-LSTM, and Transformer, was also undertaken. In addition, multivariate models, notably CFTL and CTL, were set against the two former models under different data availability scenarios for each target domain. The findings revealed that the MDS-Informer model significantly outperformed the t-SNE-Informer and PCA-Informer models, showing an improvement of approximately 23 % and 40 % in terms of MAE, respectively. Moreover, the results confirmed that the CFTL approach effectively enhanced long-term Qflow prediction accuracy compared to conventional transfer learning (CTL) and the limited data scenario (LD) without TL by approximately 30 and 47 % improvement in terms of MAE, respectively. This study demonstrates that a target domain possessing a higher similarity index with the source domain is more likely to facilitate the transfer of positive noise, thereby enhancing the predictability of the model. This phenomenon underscores the importance of domain similarity in the efficacy of TL approaches.

Dynamic video recognition for cell-encapsulating microfluidic droplets.

Droplet microfluidics is a highly sensitive and high-throughput technology extensively utilized in biomedical applications, such as single-cell sequencing and cell screening. However, its performance is highly influenced by the droplet size and single-cell encapsulation rate (following random distribution), thereby creating an urgent need for quality control. Machine learning has the potential to revolutionize droplet microfluidics, but it requires tedious pixel-level annotation for network training. This paper investigates the application software of the weakly supervised cell-counting network (WSCApp) for video recognition of microdroplets. We demonstrated its real-time performance in video processing of microfluidic droplets and further identified the locations of droplets and encapsulated cells. We verified our methods on droplets encapsulating six types of cells/beads, which were collected from various microfluidic structures. Quantitative experimental results showed that our approach can not only accurately distinguish droplet encapsulations (micro-F1 score > 0.94), but also locate each cell without any supervised location information. Furthermore, fine-tuning transfer learning on the pre-trained model also significantly reduced (>80%) annotation. This software provides a user-friendly and assistive annotation platform for the quantitative assessment of cell-encapsulating microfluidic droplets.

Brain tumor classification using fine-tuned transfer learning models on magnetic resonance imaging (MRI) images.

Brain tumors are a leading global cause of mortality, often leading to reduced life expectancy and challenging recovery. Early detection significantly improves survival rates. This paper introduces an efficient deep learning model to expedite brain tumor detection through timely and accurate identification using magnetic resonance imaging images. Our approach leverages deep transfer learning with six transfer learning algorithms: VGG16, ResNet50, MobileNetV2, DenseNet201, EfficientNetB3, and InceptionV3. We optimize data preprocessing, upsample data through augmentation, and train the models using two optimizers: Adam and AdaMax. We perform three experiments with binary and multi-class datasets, fine-tuning parameters to reduce overfitting. Model effectiveness is analyzed using various performance scores with and without cross-validation. With smaller datasets, the models achieve 100% accuracy in both training and testing without cross-validation. After applying cross-validation, the framework records an outstanding accuracy of 99.96% with a receiver operating characteristic of 100% on average across five tests. For larger datasets, accuracy ranges from 96.34% to 98.20% across different models. The methodology also demonstrates a small computation time, contributing to its reliability and speed. The study establishes a new standard for brain tumor classification, surpassing existing methods in accuracy and efficiency. Our deep learning approach, incorporating advanced transfer learning algorithms and optimized data processing, provides a robust and rapid solution for brain tumor detection.

Enhancing Learning in Fine-Tuned Transfer Learning for Rotating Machinery via Negative Transfer Mitigation

Enhancing Learning in Fine-Tuned Transfer Learning for Rotating Machinery via Negative Transfer Mitigation

Brain haemorrhage classification from CT scan images using fine-tuned transfer learning deep features

Brain haemorrhage classification from CT scan images using fine-tuned transfer learning deep features

Improving Endoscopic Image Analysis: Attention Mechanism Integration in Grid Search Fine-Tuned Transfer Learning Model for Multi-Class Gastrointestinal Disease Classification

Improving Endoscopic Image Analysis: Attention Mechanism Integration in Grid Search Fine-Tuned Transfer Learning Model for Multi-Class Gastrointestinal Disease Classification

An Optimized Uncertainty Aware Fine-Tuned Transfer Learning for COVID-19 Diagnosis from Medical Images

An Optimized Uncertainty Aware Fine-Tuned Transfer Learning for COVID-19 Diagnosis from Medical Images

A plant disease classification using one-shot learning technique with field images

Early diagnosis of plant diseases is crucial for preventing plagues and mitigating their effects on crops. The most precise automatic methods for identifying plant diseases using images of plant fields are powered by deep learning. Big image datasets should always be gathered and annotated for these methods to work, which is often not technically or financially feasible. This paper offers one-shot learning (OSL) techniques for plant disease classification with limited datasets utilizing Siamese Neural Network (SNN). There are five different crop kinds in the dataset: grape, wheat, cotton, cucumber, and corn. Five sets of images showing both healthy and diseased crops are used to represent each of the new crops. The dataset's includes 25 classes with 875 leaf images. Data augmentation techniques are used to enhance the size and dimension of the plant leaf disease image dataset. To provide effective segmentation, this paper provides a unique method for region-based image segmentation that divides an image into its most prominent regions. It also addresses issues with earlier region-based segmentation methods. SVM-based classifiers have better generalization properties as their efficiency does not depend on the number of features. Such merit is beneficial in primary diagnostics decisions to check if the input image is included in the database or not to reduce the consumed time. OSL was applied and compared to standard fine-tuning transfer learning utilizing Siamese networks and triplet loss. Siamese provides superior classification accuracy and localization accuracy with minimal errors than other approaches. The proposed approach has a total processing time of 5 ms, which makes it appropriate for real-time applications. In terms of specificity, sensitivity, precision, accuracy, MCC, and F-measure, the proposed approach beats all current machine learning algorithms for small training sets.

SEMG and IMU Data-Based Hand Gesture Recognition Method Using Multistream CNN With a Fine-Tuning Transfer Framework

sEMG and IMU Data-Based Hand Gesture Recognition Method Using Multistream CNN With a Fine-Tuning Transfer Framework

Bearing fault diagnosis method based on the Gramian angular field and an SE-ResNeXt50 transfer learning model

Fault diagnosis methods for rolling bearings based on deep learning have become a research hotspot. However, these methods mostly use convolutional neural networks (CNNs), which have the problem of gradient dispersion or disappearance as the network deepens. Moreover, directly converting vibration signals into images as network input cannot preserve the temporal correlation between signals. In the case of small datasets and complex and variable working conditions, the accuracy of fault diagnosis is low and the generalisation ability is poor. To solve the above problems, a rolling bearing fault diagnosis method based on the Gramian angular field (GAF) and an SE-ResNeXt50 transfer learning model is proposed. Firstly, the parameters of the GAF obtained from multiple experiments are selected and the one-dimensional time-series vibration signal is encoded by combining the data enhancement method, and converted into a Gramian angular difference field (GADF) diagram and a Gramian angular sum field (GASF) diagram with local time information and uniqueness. Then, a fine-tuning transfer learning strategy is used to transfer the pre-trained model parameters to an SE-ResNeXt50 model, which improves the training speed of the model and improves the overfitting problem of the model on small target datasets. Finally, the GAF diagram is used as the input to the model and a feature recalibration strategy is used to adaptively obtain the importance of each feature channel, which further improves the feature utilisation. To verify the effectiveness and superiority of the proposed method, the rolling bearing data from Case Western Reserve University are selected for experimental verification and the generalisation performance of the proposed method is tested under varying loads and different dataset scales. The results show that when there is only a small amount of data, the proposed method can still achieve high diagnosis accuracy for different loads and has better recognition accuracy and generalisation compared to other fault diagnosis methods.

Empowering COVID-19 detection: Optimizing performance through fine-tuned EfficientNet deep learning architecture

The worldwide COVID-19 pandemic has profoundly influenced the health and everyday experiences of individuals across the planet. It is a highly contagious respiratory disease requiring early and accurate detection to curb its rapid transmission. Initial testing methods primarily revolved around identifying the genetic composition of the coronavirus, exhibiting a relatively low detection rate and requiring a time-intensive procedure. To address this challenge, experts have suggested using radiological imagery, particularly chest X-rays, as a valuable approach within the diagnostic protocol. This study investigates the potential of leveraging radiographic imaging (X-rays) with deep learning algorithms to swiftly and precisely identify COVID-19 patients. The proposed approach elevates the detection accuracy by fine-tuning with appropriate layers on various established transfer learning models. The experimentation was conducted on a COVID-19 X-ray dataset containing 2000 images. The accuracy rates achieved were impressive of 99.55%, 97.32%, 99.11%, 99.55%, 99.11% and 100% for Xception, InceptionResNetV2, ResNet50 , ResNet50V2, EfficientNetB0 and EfficientNetB4 respectively. The fine-tuned EfficientNetB4 achieved an excellent accuracy score, showcasing its potential as a robust COVID-19 detection model. Furthermore, EfficientNetB4 excelled in identifying Lung disease using Chest X-ray dataset containing 4,350 Images, achieving remarkable performance with an accuracy of 99.17%, precision of 99.13%, recall of 99.16%, and f1-score of 99.14%. These results highlight the promise of fine-tuned transfer learning for efficient lung detection through medical imaging, especially with X-ray images. This research offers radiologists an effective means of aiding rapid and precise COVID-19 diagnosis and contributes valuable assistance for healthcare professionals in accurately identifying affected patients.