Binary Cross-entropy Loss Research Articles

Ai-aided diagnosis of oral X-ray images of periapical films based on deep learning

Oral X-ray images provide a useful technical means by which dentists examine teeth for dental problems, but the diagnostic process is defective due to its over-reliance on dentists’ subjective judgments, lack of objective criteria, etc. In this context, this study examined the AI-aided diagnosis of periapical films based on deep learning..Based on YOLOv7-X, a YOLO-DENTAL network architecture was used to detect dental caries, dental defects, periapical lesions, and coronal restorations in periapical films. Firstly, the coordinate attention (CA) mechanism was introduced into the backbone feature extraction network, and a backbone-CA structure was presented to enhance the feature extraction capability of the network. Secondly, a simplified Bi-FPN structure was put forward and applied to the feature fusion part of the network to effectively improve its multi-scale feature fusion effect. Thirdly, the existing anchor-based detection head was replaced by an anchor-free decoupled head to simplify operational parameters while improving the generalized detection capability of the model over lesion regions. In the loss function part, existing CIoU loss was replaced by SIoU loss, a border loss function containing direction information. The focal loss containing a weight factor was introduced in calculating confidence loss as a substitute for the existing binary cross entropy loss function to balance positive and negative samples. Meanwhile, a study of ablation experiment was completed. The results validated the positive gain effect of each optimization strategy on the model. The final YOLO-DENTAL network structure exhibited an mAP value of 86.81%, higher than that of YOLOv7-X (79.95%). The effect of aided diagnosis was well achieved.

MicroSegNet: A deep learning approach for prostate segmentation on micro-ultrasound images

Micro-ultrasound (micro-US) is a novel 29-MHz ultrasound technique that provides 3-4 times higher resolution than traditional ultrasound, potentially enabling low-cost, accurate diagnosis of prostate cancer. Accurate prostate segmentation is crucial for prostate volume measurement, cancer diagnosis, prostate biopsy, and treatment planning. However, prostate segmentation on micro-US is challenging due to artifacts and indistinct borders between the prostate, bladder, and urethra in the midline. This paper presents MicroSegNet, a multi-scale annotation-guided transformer UNet model designed specifically to tackle these challenges. During the training process, MicroSegNet focuses more on regions that are hard to segment (hard regions), characterized by discrepancies between expert and non-expert annotations. We achieve this by proposing an annotation-guided binary cross entropy (AG-BCE) loss that assigns a larger weight to prediction errors in hard regions and a lower weight to prediction errors in easy regions. The AG-BCE loss was seamlessly integrated into the training process through the utilization of multi-scale deep supervision, enabling MicroSegNet to capture global contextual dependencies and local information at various scales. We trained our model using micro-US images from 55 patients, followed by evaluation on 20 patients. Our MicroSegNet model achieved a Dice coefficient of 0.939 and a Hausdorff distance of 2.02 mm, outperforming several state-of-the-art segmentation methods, as well as three human annotators with different experience levels. Our code is publicly available at https://github.com/mirthAI/MicroSegNet and our dataset is publicly available at https://zenodo.org/records/10475293.

Transformer Fusion and Pixel-Level Contrastive Learning for RGB-D Salient Object Detection

Current RGB-D salient object detection (RGB-D SOD) methods mainly develop a generalizable model trained by binary cross-entropy (BCE) loss based on convolutional or Transformer backbones. However, they usually exploit convolutional modules to fuse multi-modality features, with little attention paid to capturing the long-range multi-modality interactions for feature fusion. Furthermore, BCE loss does not explicitly explore intra- and inter-pixel relationships in a joint embedding space. To address these issues, we propose a cross-modality interaction parallel-transformer (CIPT) module, which better captures the long-range multi-modality interactions, generating more comprehensive fusion features. Besides, we propose a pixel-level contrastive learning (PCL) method that improves inter-pixel discrimination and intra-pixel compactness, resulting in a well-structured embedding space and a better saliency detector. Specifically, we propose an asymmetric network (TPCL) for RGB-D SOD, which consists of a Swin V2 Transformer-based backbone and a designed lightweight backbone (LDNet). Moreover, an edge-guided module and a feature enhancement (FE) module are proposed to refine the learned fusion features. Extensive experiments demonstrate that our method achieves excellent performance against 15 state-of-the-art methods on seven public datasets. We expect our work to facilitate the exploration of applying Transformer and contrastive learning for RGB-D SOD tasks. Our codes and predicted saliency maps will be released on GitHub <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/TomorrowJW/TPCL_RGBDSOD</uri> .

COVLIAS 3.0: cloud-based quantized hybrid UNet3+ deep learning for COVID-19 lesion detection in lung computed tomography.

When RT-PCR is ineffective in early diagnosis and understanding of COVID-19 severity, Computed Tomography (CT) scans are needed for COVID diagnosis, especially in patients having high ground-glass opacities, consolidations, and crazy paving. Radiologists find the manual method for lesion detection in CT very challenging and tedious. Previously solo deep learning (SDL) was tried but they had low to moderate-level performance. This study presents two new cloud-based quantized deep learning UNet3+ hybrid (HDL) models, which incorporated full-scale skip connections to enhance and improve the detections. Annotations from expert radiologists were used to train one SDL (UNet3+), and two HDL models, namely, VGG-UNet3+ and ResNet-UNet3+. For accuracy, 5-fold cross-validation protocols, training on 3,500 CT scans, and testing on unseen 500 CT scans were adopted in the cloud framework. Two kinds of loss functions were used: Dice Similarity (DS) and binary cross-entropy (BCE). Performance was evaluated using (i) Area error, (ii) DS, (iii) Jaccard Index, (iii) Bland-Altman, and (iv) Correlation plots. Among the two HDL models, ResNet-UNet3+ was superior to UNet3+ by 17 and 10% for Dice and BCE loss. The models were further compressed using quantization showing a percentage size reduction of 66.76, 36.64, and 46.23%, respectively, for UNet3+, VGG-UNet3+, and ResNet-UNet3+. Its stability and reliability were proved by statistical tests such as the Mann-Whitney, Paired t-Test, Wilcoxon test, and Friedman test all of which had a p < 0.001. Full-scale skip connections of UNet3+ with VGG and ResNet in HDL framework proved the hypothesis showing powerful results improving the detection accuracy of COVID-19.

LST-AI: A deep learning ensemble for accurate MS lesion segmentation

Automated segmentation of brain white matter lesions is crucial for both clinical assessment and scientific research in multiple sclerosis (MS). Over a decade ago, we introduced an engineered lesion segmentation tool, LST. While recent lesion segmentation approaches have leveraged artificial intelligence (AI), they often remain proprietary and difficult to adopt. As an open-source tool, we present LST-AI, an advanced deep learning-based extension of LST that consists of an ensemble of three 3D U-Nets.LST-AI explicitly addresses the imbalance between white matter (WM) lesions and non-lesioned WM. It employs a composite loss function incorporating binary cross-entropy and Tversky loss to improve segmentation of the highly heterogeneous MS lesions. We train the network ensemble on 491 MS pairs of T1-weighted and FLAIR images, collected in-house from a 3T MRI scanner, and expert neuroradiologists manually segmented the utilized lesion maps for training. LST-AI also includes a lesion location annotation tool, labeling lesions as periventricular, infratentorial, and juxtacortical according to the 2017 McDonald criteria, and, additionally, as subcortical. We conduct evaluations on 103 test cases consisting of publicly available data using the Anima segmentation validation tools and compare LST-AI with several publicly available lesion segmentation models.Our empirical analysis shows that LST-AI achieves superior performance compared to existing methods. Its Dice and F1 scores exceeded 0.62, outperforming LST, SAMSEG (Sequence Adaptive Multimodal SEGmentation), and the popular nnUNet framework, which all scored below 0.56. Notably, LST-AI demonstrated exceptional performance on the MSSEG-1 challenge dataset, an international WM lesion segmentation challenge, with a Dice score of 0.65 and an F1 score of 0.63—surpassing all other competing models at the time of the challenge. With increasing lesion volume, the lesion detection rate rapidly increased with a detection rate of >75% for lesions with a volume between 10 mm3 and 100 mm3. Given its higher segmentation performance, we recommend that research groups currently using LST transition to LST-AI. To facilitate broad adoption, we are releasing LST-AI as an open-source model, available as a command-line tool, dockerized container, or Python script, enabling diverse applications across multiple platforms.

A colour vision test assisted by neural networks

Aims/Purpose: Traditional computerized colour vision tests resort to the magnitude of chromatic thresholds to determine the quality of the colour vision of an observer. The limits of perceived colours are tested against a background that will be conspicuous to normal colour vision observers, but inconspicuous to colour vision deficient (CVD) observers. The purpose of this work was to implement a Neural Network capable of performing independent and automatic classification of the colour vision of an observer, using as inputs the sRGB chromaticity coordinates, response times, and eye movements obtained using a colour vision test on a computer screen calibrated in colour and luminance (Display++).Methods: Chromatic discrimination limits were estimated using a staircase procedure that tested colour vision close to the dichromatic confusion lines. Only the last 8 responses were considered in the analysis. 50 normal colour vision observers and 8 CVD observers performed the test on Display++. CVD participants were diagnosed using the results of the anomaloscope and the CAD colour vision tests. A Sequential Neural Network was implemented with a dense structure (256–128–64‐16‐1 nodes), with relu and sigmoid activation functions, dropout commands to prevent overfitting, the Adam optimizer, and the binary crossentropy loss function. The performance of the neural network was estimated by computing its accuracy, where variables are related to the results obtained after training the Neural Network and inspecting the test results. The database was divided into a training/test following different splitting strategies: T1 observers' responses considering sRGB coordinates, response times, and eye movements. T2 observers' responses considering only sRGB coordinates.Results: The averages and standard deviations of the estimated accuracy for five consecutive trials with a random division between training/test for conditions T1 and T2 was 91% ± 7% and 93% ± 5% respectively. It was found that the implemented neural network had an accuracy of more than 90% across all conditions.Conclusions: The accuracies found across conditions considered seem to indicate that the implemented Neural Network can analyse and classify the colour vision of human observers, with an accuracy greater than 90% of success.

CCN+: A neuro-symbolic framework for deep learning with requirements

For their outstanding ability of finding hidden patterns in data, deep learning models have been extensively applied in many different domains. However, recent works have shown that, if a set of requirements expressing inherent knowledge about the problem at hand is given, then neural networks often fail to comply with them. This represents a major drawback for deep learning models, as requirements compliance is normally considered a necessary condition for standard software deployment. In this paper, we propose a novel neuro-symbolic framework able to make any neural network compliant by design to a given set of requirements over the output space expressed in full propositional logic. This framework, called CCN+, integrates the requirements into the output layer of the neural network by applying multiple inference rules that ensure compliance with the requirements and adapts the standard binary cross-entropy loss function to the requirement output layer. As a result, not only the outputted predictions are guaranteed to be compliant with the requirements, but the neural network itself learns how to exploit the domain knowledge expressed by the requirements to get better performance. We conduct an extensive experimental evaluation of CCN+ on 19 real-world multi-label classification datasets with propositional logic requirements, including a challenging dataset for autonomous driving. Our experimental analysis confirms that CCN+ is able to outperform both its neural counterparts and the state-of-the-art models.

CIDN: A context interactive deep network with edge-aware for X-ray angiography images segmentation

Accurate segmentation of X-ray angiography images is imperative for cardiovascular disease diagnosis. Despite significant strides in segmentation through deep learning, challenges persist in precisely delineating vessel edges. This paper introduces a Context Interactive Deep Network (CIDN) tailored for coronary vessel segmentation. CIDN integrates a Bio-inspired Attention Block (BAB) and a Multi-scale Interactive Block (MIB) to optimize the capture of spatial and edge information. The encoder-decoder facilitates optimal interaction between low-level features and high-level semantics. A compound loss function, amalgamating binary cross-entropy and active contour elasticity loss, enhances the recognition of intricate vessel edges. In experiments conducted on both public and private X-ray contrast images, the CIDN exhibits superior performance compared to the state-of-the-art Spatial Multi-scale Attention U-improved Network (SMAU-Net). Specifically, on dataset DCA1, the F1 value reached 0.7675, with an accuracy of 0.9795. Furthermore, on dataset JMA, the CIDN model achieved an F1 value of 0.8732, coupled with an accuracy of 0.9757. Accurate segmentation provides clinicians with precise vessel depictions, aiding in the identification of coronary stenosis and facilitating more informed diagnostic and therapeutic decisions.

Optimal Tilt-Wing eVTOL Takeoff Trajectory Prediction Using Regression Generative Adversarial Networks

Electric vertical takeoff and landing (eVTOL) aircraft have attracted tremendous attention nowadays due to their flexible maneuverability, precise control, cost efficiency, and low noise. The optimal takeoff trajectory design is a key component of cost-effective and passenger-friendly eVTOL systems. However, conventional design optimization is typically computationally prohibitive due to the adoption of high-fidelity simulation models in an iterative manner. Machine learning (ML) allows rapid decision making; however, new ML surrogate modeling architectures and strategies are still desired to address large-scale problems. Therefore, we showcase a novel regression generative adversarial network (regGAN) surrogate for fast interactive optimal takeoff trajectory predictions of eVTOL aircraft. The regGAN leverages generative adversarial network architectures for regression tasks with a combined loss function of a mean squared error (MSE) loss and an adversarial binary cross-entropy (BC) loss. Moreover, we introduce a surrogate-based inverse mapping concept into eVTOL optimal trajectory designs for the first time. In particular, an inverse-mapping surrogate takes design requirements (including design constraints and flight condition parameters) as input and directly predicts optimal trajectory designs, with no need to run design optimizations once trained. We demonstrated the regGAN on optimal takeoff trajectory designs for the Airbus A3 Vahana. The results revealed that regGAN outperformed reference surrogate strategies, including multi-output Gaussian processes and conditional generative adversarial network surrogates, by matching simulation-based ground truth with 99.6% relative testing accuracy using 1000 training samples. A parametric study showed that a regGAN surrogate with an MSE weight of one and a BC weight of 0.01 consistently achieved over 99.5% accuracy (denoting negligible predictive errors) using 400 training samples, while other regGAN models require at least 800 samples.

Stance-Based Fake News Identification on Social Media with Hybrid CNN and RNN-LSTM Models

Today, fake news can be readily generated and disseminated via social media platforms. Misinformation and hoaxes propagated via online social media or traditional news media are commonly referred to as fake news. Stance-based fake news is based on the opinions of an audience rather than providing correct facts. This study presents a hybrid model focusing on the CNN model and RNN-LSTM model to identify fake news. A balanced dataset of 216k news items called ‘SherLock-FakeNewsNet’ is explored throughout the study resulting in the proposed hybrid model. The NLTK toolkit is utilized to perform some noise reduction. After that, the Keras tokenizer and pre-processor are used for the tokenization and pre-processing steps. Next, the text data is fitted and represented using pre-trained GloVe word embeddings. CNN model is applied with Conv1D layers and MaxPooling layers to extract higher-level features of the text. Following this, to detect the longer context of the given text and to capture interdependencies among word sequences LSTM units are employed in the RNN model. Dropout layers are carefully chosen to reduce overfitting in the final hybrid model. The proposed hybrid model achieves the highest accuracy rate of 92% by outperforming most of the conventional models today with an average Precision of 91%, Recall of 91%, and F1-Score of 91% using an Adam optimizer and a Binary Cross-Entropy loss function. Based on the five experiments performed in the study, we have presented the results by comparing our proposed hybrid model with five related datasets.

SENTIMENT ANALYSIS USING LONG TERM MEMORY (LSTM) BOOK CASE STUDY: UNSOLICITED ADVICE FOR MURDERERS BY VERA WONG'S

Reading books is one of the most effective ways to reduce stress. In today's digital era, access to finding and buying books is getting easier, so reader reviews are important in choosing books that match interests. However, with a large number of reviews, Natural Language Processing (NLP) with the Long Short-Term Memory (LSTM) method is used to help analyze positive and negative sentiments from many book reviews. This sentiment analysis is useful for readers to evaluate the quality of books, as well as for authors and sellers to find out the opinions of readers and improve the quality of their work. In this study, the book review dataset "Vera Wong's Unsolicited Advice for Murderers" from the Goodreads website is used, which is then divided into training data and validation data with a ratio of 75%: 25%. The Long Short-Term Memory (LSTM) method is used to analyze the sentiment of the reviews. The model architecture built consists of Embedding Layer, LSTM Layer with 128 neuron units, 3 Dense Layer with ReLU activation function, 3 Dropout Layer, and Fully Connected Layer with and Sigmoid activation function, Binary Cross Entropy loss function, and RMSprop optimizer. The model training process was conducted with 30 epochs. The evaluation results show that the model achieved an accuracy of 90%, indicating the model performs relatively well in correctly classifying positive sentiments.

DualAttNet: Synergistic fusion of image-level and fine-grained disease attention for multi-label lesion detection in chest X-rays

Chest radiographs are the most commonly performed radiological examinations for lesion detection. Recent advances in deep learning have led to encouraging results in various thoracic disease detection tasks. Particularly, the architecture with feature pyramid network performs the ability to recognise targets with different sizes. However, such networks are difficult to focus on lesion regions in chest X-rays due to their high resemblance in vision. In this paper, we propose a dual attention supervised module for multi-label lesion detection in chest radiographs, named DualAttNet. It efficiently fuses global and local lesion classification information based on an image-level attention block and a fine-grained disease attention algorithm. A binary cross entropy loss function is used to calculate the difference between the attention map and ground truth at image level. The generated gradient flow is leveraged to refine pyramid representations and highlight lesion-related features. We evaluate the proposed model on VinDr-CXR, ChestX-ray8 and COVID-19 datasets. The experimental results show that DualAttNet surpasses baselines by 0.6% to 2.7% mAP and 1.4% to 4.7% AP50 with different detection architectures. The code for our work and more technical details can be found at https://github.com/xq141839/DualAttNet.

Enhancing Label Correlations in multi-label classification through global-local label specific feature learning to Fill Missing labels

In multi-label classification, challenges arise from missing labels due to subjective analysis or label ambiguity. This makes it difficult to accurately capture label correlations and enhance classifier performance. Previous research has primarily focused on obtaining label correlations with sparsity assumptions from incomplete label matrices to fill in these missing labels. However, this approach leads to inaccurate label correlations since it relies on incomplete label matrix information. To overcome this issue and improve accuracy, this paper proposes a new method called LCFM, which effectively learns label correlations using a combination of global and local label-specific features to recover missing labels. Firstly, we rely on the sparsity assumption of Lasso regression coefficients to select label-specific features from the training samples. Secondly, we introduce l2-norm regularization to learn global label correlations through global label-specific features. simultaneously, we utilize similar label-specific features to describe local label correlations. Finally, we minimize the difference between predicted results and true labels using the binary cross-entropy loss function and employ the Accelerated Proximal Gradient (APG) algorithm for model optimization. Experimental results comparing LCFM with five benchmark indicators on nine datasets demonstrate that our proposed method outperforms other advanced methods, showing strong competitiveness.

MLMQ-IR: Multi-label multi-query image retrieval based on the variance of Hamming distance

Image retrieval (IR) methods extract the most relevant images to the query images from an image database. The existing IR methods, which retrieve images with a low degree of similarity, use computationally expensive approaches. In this study, a novel Multi-Label Multi-Query IR (MLMQ-IR) method based on the variance of Hamming distance is presented for the query of multiple images having multiple labels. The MLMQ-IR uses deep learning-based hashing code generation with ResNet50 structure. The variance evaluates the minimum variation of the distances between the query and database images, providing the trade-off images according to the center of Pareto space. Moreover, the MLMQ-IR exploits a new Triple Loss Multi-Label Hashing (TLMH) depending on binary cross-entropy loss and bit-balance loss functions. The MLMQ-IR is compared with recent multi-label and multi-query methods through MIRFLICKR-25 K, MS-COCO and NUS-WIDE datasets in terms of three well-known metrics, and the methods are ranked with succussed sorting. As a result, the MLMQ-IR method has the best avg. mean rank with 1.86. The results manifest that the MLMQ-IR provides the most similar retrieved images to the query images owing to utilizing the variance which is the efficient and fast IR approach.

NIMG-57. PERFORMANCE OF DEEP LEARNING IN MGMT PROMOTER METHYLATION STATUS PREDICTION USING BRAIN MRI: RESULTS FROM A LARGE COHORT OF IDH-WILDTYPE GLIOMAS TESTED BY A SINGLE METHYLATION ASSAY

Abstract There have been several attempts to predict the MGMT promoter methylation status of gliomas using deep learning (DL), with mixed results. These studies did not account for IDH mutation status and did not utilize a uniform methylation assay, introducing potential confounders in the experimental design. We assessed the performance of DL applied to brain MRI to predict the MGMT promoter methylation status of IDH-wildtype gliomas tested with a standardized assay. We included 539 IDH-wildtype gliomas tested with real-time methylation-specific PCR from 2017-2022; 205 (38%) were MGMT promoter methylated. The assay targets 8 downstream MGMT CpG sites using ACTB as control; methylation status is determined based on the difference of PCR crossing-point values: methylated if difference &amp;lt; 5, unmethylated if difference &amp;gt; 7, indeterminate otherwise. We used post-contrast T1 and T2 MRIs as inputs for our models trained using stratified 5-fold cross-validation, 400 epochs, batch size of 16, a learning rate scheduler starting at 10-3, weighted binary cross-entropy loss, and AdamW optimizer. We repeated the experiments using two 3D DL architectures (DenseNet121 and EfficientNetB0) and two image preprocessing pipelines: the BraTS protocol and one that uses padding instead of registration to an MRI atlas, keeping constant other experimental variables. For Densenet121, the best cross-validation AUROCC and F1-score (median, median absolute deviation) were 0.67 (0.06), 0.59 (0.02) using the BraTS protocol, and 0.67 (0.06), 0.58 (0.01) using padding. For EfficientNetB0, were 0.67 (0.02), 0.66 (0.01) using the BraTS protocol and 0.67 (0.01), 0.68 (0.01) using padding. Our findings suggest that it is unlikely that DL models trained on brain MRI can predict the MGMT promoter methylation of IDH-wildtype gliomas. Importantly, our study is the first to focus on an IDH-wildtype adult diffuse glioma cohort tested with a single, standardized MGMT methylation assay, eliminating the variability introduced by different assays and IDH status.

CA-YOLOX: Deep Learning-Guided Road Intersection Location From High-Resolution Remote Sensing Images

The location of road intersection from high resolution remote sensing (HRRS) images can be automatically obtained by deep learning. This has become one of the current data sources in urban smart transportation. However, limited by the small size, diverse types, complex distribution, and missing sample labels of road intersections in actual scenarios, it is difficult to accurately represent key features of road intersection by deep neural network (DNN) model. A new coordinate attention (CA) module-YOLOX (CA-YOLOX) method for accurately locating road intersections from HRRS images is presented. First, the spatial pyramid pooling (SPP) module is introduced into the backbone convolution network between the Darknet-53’ last feature layer and feature pyramid networks (FPN) structure. Second, the CA module is embedded into the feature fusion structure in FPN to focus more on the spatial shape distribution and texture features of road intersections. Third, we use focal loss to replace the traditional binary cross entropy (BCE) loss in the confidence loss to improve the iteration speed of the CA-YOLOX network. Finally, an extensive empirical experiment on Potsdam, IKONOS datasets, and ablation study is then implemented and tested. The results show that the presented CA-YOLOX method can promote the location accuracy of road intersection from HRRS images compared to the traditional You only look once (YOLO) model.

Crack segmentation of imbalanced data: The role of loss functions

Loss functions, which govern a deep learning-based optimization process, have been widely investigated to handle the class imbalanced data issue in crack segmentation. However, their performance varies according to models and datasets, making it challenging to choose the most appropriate ones. To address this issue, the paper conducts a large-scale performance comparison of twelve commonly used loss functions on four benchmark datasets. Various aspects are considered using a statistical test-based ranking scheme, which integrates accuracy, sensitivity to threshold change, and varying degrees of imbalance severity for a comprehensive comparison. The results show that most loss functions achieve relatively similar accuracy on mildly imbalanced datasets, while weighted binary cross-entropy loss, Focal loss, Dice-based loss, and compound loss functions significantly outperform others as imbalance severity increases. In general, Focal Tversky loss function exhibits excellent performance in handling the imbalanced data issue.

Artificial-Neural-Network-Driven Innovations in Time-Varying Process Diagnosis of Low-K Oxide Deposition.

To address the challenges in real-time process diagnosis within the semiconductor manufacturing industry, this paper presents a novel machine learning approach for analyzing the time-varying 10th harmonics during the deposition of low-k oxide (SiOF) on a 600 Å undoped silicate glass thin liner using a high-density plasma chemical vapor deposition system. The 10th harmonics, which are high-frequency components 10 times the fundamental frequency, are generated in the plasma sheath because of their nonlinear nature. An artificial neural network with a three-hidden-layer architecture was applied and optimized using k-fold cross-validation to analyze the harmonics generated in the plasma sheath during the deposition process. The model exhibited a binary cross-entropy loss of 0.1277 and achieved an accuracy of 0.9461. This approach enables the accurate prediction of process performance, resulting in significant cost reduction and enhancement of semiconductor manufacturing processes. This model has the potential to improve defect control and yield, thereby benefiting the semiconductor industry. Despite the limitations imposed by the limited dataset, the model demonstrated promising results, and further performance improvements are anticipated with the inclusion of additional data in future studies.

Siamese Networks for Speaker Identification on Resource-Constrained Platforms

This paper investigates the implementation of a lightweight Siamese neural network for enhancing speaker identification accuracy and inference speed in embedded systems. Integrating speaker identification into embedded systems can improve portability and versatility. Siamese neural networks achieve speaker identification by comparing input voice samples to reference voices in a database, effectively extracting features and classifying speakers accurately. Considering the trade-off between accuracy and complexity, as well as hardware constraints in embedded systems, various neural networks could be applied to speaker identification. This paper compares the incorporation of CNN architectures targeted for embedded systems, MCUNet, SqueezeNet and MobileNetv2, to implement Siamese neural networks on a Raspberry Pi. Our experiments demonstrate that MCUNet achieves 85% accuracy with a 0.23-second inference time. In comparison, the larger MobileNetv2 attains 84.5% accuracy with a 0.32-second inference time. Additionally, contrastive loss was superior to binary cross-entropy loss in the Siamese neural network. The system using contrastive loss had almost 68% lower loss scores, resulting in a more stable performance and more accurate predictions. In conclusion, this paper establishes that an appropriate lightweight Siamese neural network, combined with contrastive loss, can significantly improve speaker identification accuracy, and enable efficient deployment on resource-constrained platforms.

Multi-Label Classification of Chest X-ray Abnormalities Using Transfer Learning Techniques.

In recent years, deep neural networks have enabled countless innovations in the field of image classification. Encouraged by success in this field, researchers worldwide have demonstrated how to use Convolutional Neural Network techniques in medical imaging problems. In this article, the results were obtained through the use of the EfficientNet in the task of classifying 14 different diseases based on chest X-ray images coming from the NIH (National Institutes of Health) ChestX-ray14 dataset. The approach addresses dataset imbalances by introducing a custom split to ensure fair representation. Binary cross entropy loss is utilized to handle the multi-label difficulty. The model architecture comprises an EfficientNet backbone for feature extraction, succeeded by sequential layers including GlobalAveragePooling, Dense, and BatchNormalization. The main contribution of this paper is a proposed solution that outperforms previous state-of-the-art deep learning models average area under the receiver operating characteristic curve-AUC-ROC (score: 84.28%). The usage of the transfer-learning technique and traditional deep learning engineering techniques was shown to enable us to obtain such results on consumer-class GPUs (graphics processing units).