MRGazer: decoding eye gaze points from functional magnetic resonance imaging in individual space.

Eye-tracking research has proven valuable in understanding numerous cognitive functions. Recently, Frey et al. provided an exciting deep learning method for learning eye movements from functional magnetic resonance imaging (fMRI) data. It employed the multi-step co-registration of fMRI into the group template to obtain eyeball signal, and thus required additional templates and was time consuming. To resolve this issue, in this paper, we propose a framework named MRGazer for predicting eye gaze points from fMRI in individual space. The MRGazer consists of an eyeball extraction module and a residual network-based eye gaze prediction module. Compared to the previous method, the proposed framework skips the fMRI co-registration step, simplifies the processing protocol, and achieves end-to-end eye gaze regression. The proposed method achieved superior performance in eye fixation regression (Euclidean error, EE=2.04°) than the co-registration-based method (EE=2.89°)， and delivered objective results within a shorter time (~0.02 second/volume) than prior method (~0.3 second/volume). The code is available at https://github.com/ustc-bmec/MRGazer.

Scoring facial attractiveness with deep convolutional neural networks: How training on standardized images reduces the bias of facial expressions.

OBJECTIVE: In many medical disciplines, facial attractiveness is part of the diagnosis, yet its scoring might be confounded by facial expressions. The intent was to apply deep convolutional neural networks (CNN) to identify how facial expressions affect facial attractiveness and to explore whether a dedicated training of the CNN is able to reduce the bias of facial expressions. MATERIALS AND METHODS: Frontal facial images (n = 840) of 40 female participants (mean age 24.5 years) were taken adapting a neutral facial expression and the six universal facial expressions. Facial attractiveness was computed by means of a face detector, deep convolutional neural networks, standard support vector regression for facial beauty, visual regularized collaborative filtering and a regression technique for handling visual queries without rating history. CNN was first trained on random facial photographs from a dating website and then further trained on the Chicago Face Database (CFD) to increase its suitability to medical conditions. Both algorithms scored every image for attractiveness. RESULTS: Facial expressions affect facial attractiveness scores significantly. Scores from CNN additionally trained on CFD had less variability between the expressions (range 54.3-60.9 compared to range: 32.6-49.5) and less variance within the scores (P ≤ .05), but also caused a shift in the ranking of the expressions' facial attractiveness. CONCLUSION: Facial expressions confound attractiveness scores. Training on norming images generated scores less susceptible to distortion, but more difficult to interpret. Scoring facial attractiveness based on CNN seems promising, but AI solutions must be developed on CNN trained to recognize facial expressions as distractors.

Adaptive habitat biogeography-based optimizer for optimizing deep CNN hyperparameters in image classification.

Deep Convolutional Neural Networks (DCNNs) have shown remarkable success in image classification tasks, but optimizing their hyperparameters can be challenging due to their complex structure. This paper develops the Adaptive Habitat Biogeography-Based Optimizer (AHBBO) for tuning the hyperparameters of DCNNs in image classification tasks. In complicated optimization problems, the BBO suffers from premature convergence and insufficient exploration. In this regard, an adaptable habitat is presented as a solution to these problems; it would permit variable habitat sizes and regulated mutation. Better optimization performance and a greater chance of finding high-quality solutions across a wide range of problem domains are the results of this modification's increased exploration and population diversity. AHBBO is tested on 53 benchmark optimization functions and demonstrates its effectiveness in improving initial stochastic solutions and converging faster to the optimum. Furthermore, DCNN-AHBBO is compared to 23 well-known image classifiers on nine challenging image classification problems and shows superior performance in reducing the error rate by up to 5.14%. Our proposed algorithm outperforms 13 benchmark classifiers in 87 out of 95 evaluations, providing a high-performance and reliable solution for optimizing DNNs in image classification tasks. This research contributes to the field of deep learning by proposing a new optimization algorithm that can improve the efficiency of deep neural networks in image classification.

Cancerous and Non-Cancerous MRI Classification Using Dual DCNN Approach.

Brain cancer is a life-threatening disease requiring close attention. Early and accurate diagnosis using non-invasive medical imaging is critical for successful treatment and patient survival. However, manual diagnosis by radiologist experts is time-consuming and has limitations in processing large datasets efficiently. Therefore, efficient systems capable of analyzing vast amounts of medical data for early tumor detection are urgently needed. Deep learning (DL) with deep convolutional neural networks (DCNNs) emerges as a promising tool for understanding diseases like brain cancer through medical imaging modalities, especially MRI, which provides detailed soft tissue contrast for visualizing tumors and organs. DL techniques have become more and more popular in current research on brain tumor detection. Unlike traditional machine learning methods requiring manual feature extraction, DL models are adept at handling complex data like MRIs and excel in classification tasks, making them well-suited for medical image analysis applications. This study presents a novel Dual DCNN model that can accurately classify cancerous and non-cancerous MRI samples. Our Dual DCNN model uses two well-performed DL models, i.e., inceptionV3 and denseNet121. Features are extracted from these models by appending a global max pooling layer. The extracted features are then utilized to train the model with the addition of five fully connected layers and finally accurately classify MRI samples as cancerous or non-cancerous. The fully connected layers are retrained to learn the extracted features for better accuracy. The technique achieves 99%, 99%, 98%, and 99% of accuracy, precision, recall, and f1-scores, respectively. Furthermore, this study compares the Dual DCNN's performance against various well-known DL models, including DenseNet121, InceptionV3, ResNet architectures, EfficientNetB2, SqueezeNet, VGG16, AlexNet, and LeNet-5, with different learning rates. This study indicates that our proposed approach outperforms these established models in terms of performance.

Explainable DCNN Decision Framework for Breast Lesion Classification from Ultrasound Images Based on Cancer Characteristics.

In recent years, deep convolutional neural networks (DCNNs) have shown promising performance in medical image analysis, including breast lesion classification in 2D ultrasound (US) images. Despite the outstanding performance of DCNN solutions, explaining their decisions remains an open investigation. Yet, the explainability of DCNN models has become essential for healthcare systems to accept and trust the models. This paper presents a novel framework for explaining DCNN classification decisions of lesions in ultrasound images using the saliency maps linking the DCNN decisions to known cancer characteristics in the medical domain. The proposed framework consists of three main phases. First, DCNN models for classification in ultrasound images are built. Next, selected methods for visualization are applied to obtain saliency maps on the input images of the DCNN models. In the final phase, the visualization outputs and domain-known cancer characteristics are mapped. The paper then demonstrates the use of the framework for breast lesion classification from ultrasound images. We first follow the transfer learning approach and build two DCNN models. We then analyze the visualization outputs of the trained DCNN models using the EGrad-CAM and Ablation-CAM methods. We map the DCNN model decisions of benign and malignant lesions through the visualization outputs to the characteristics such as echogenicity, calcification, shape, and margin. A retrospective dataset of 1298 US images collected from different hospitals is used to evaluate the effectiveness of the framework. The test results show that these characteristics contribute differently to the benign and malignant lesions' decisions. Our study provides the foundation for other researchers to explain the DCNN classification decisions of other cancer types.

Involving logical clinical knowledge into deep neural networks to improve bladder tumor segmentation.

Segmentation of bladder tumors from medical radiographic images is of great significance for early detection, diagnosis and prognosis evaluation of bladder cancer. Deep Convolution Neural Networks (DCNNs) have been successfully used for bladder tumor segmentation, but the segmentation based on DCNN is data-hungry for model training and ignores clinical knowledge. From the clinical view, bladder tumors originate from the mucosal surface of bladder and must rely on the bladder wall to survive and grow. This clinical knowledge of tumor location is helpful to improve the bladder tumor segmentation. To achieve this, we propose a novel bladder tumor segmentation method, which incorporates the clinical logic rules of bladder tumor and bladder wall into DCNNs to harness the tumor segmentation. Clinical logical rules provide a semantic and human-readable knowledge representation and are easy for knowledge acquisition from clinicians. In addition, incorporating logical rules of clinical knowledge helps to reduce the data dependency of the segmentation network, and enables precise segmentation results even with limited number of annotated images. Experiments on bladder MR images collected from the collaborating hospital validate the effectiveness of the proposed bladder tumor segmentation method.

A novel approach to recognition of Alzheimer's and Parkinson's diseases: random subspace ensemble classifier based on deep hybrid features with a super-resolution image.

Background: Artificial intelligence technologies have great potential in classifying neurodegenerative diseases such as Alzheimer's and Parkinson's. These technologies can aid in early diagnosis, enhance classification accuracy, and improve patient access to appropriate treatments. For this purpose, we focused on AI-based auto-diagnosis of Alzheimer's disease, Parkinson's disease, and healthy MRI images. Methods: In the current study, a deep hybrid network based on an ensemble classifier and convolutional neural network was designed. First, a very deep super-resolution neural network was adapted to improve the resolution of MRI images. Low and high-level features were extracted from the images processed with the hybrid deep convolutional neural network. Finally, these deep features are given as input to the k-nearest neighbor (KNN)-based random subspace ensemble classifier. Results: A 3-class dataset containing publicly available MRI images was utilized to test the proposed architecture. In experimental works, the proposed model produced 99.11% accuracy, 98.75% sensitivity, 99.54% specificity, 98.65% precision, and 98.70% F1-score performance values. The results indicate that our AI system has the potential to provide valuable diagnostic assistance in clinical settings.

Deep learning from atrioventricular plane displacement in patients with Takotsubo syndrome: lighting up the black-box.

Aims: The spatiotemporal deep convolutional neural network (DCNN) helps reduce echocardiographic readers' erroneous 'judgement calls' on Takotsubo syndrome (TTS). The aim of this study was to improve the interpretability of the spatiotemporal DCNN to discover latent imaging features associated with causative TTS pathophysiology. Methods and results: We applied gradient-weighted class activation mapping analysis to visualize an established spatiotemporal DCNN based on the echocardiographic videos to differentiate TTS (150 patients) from anterior wall ST-segment elevation myocardial infarction (STEMI, 150 patients). Forty-eight human expert readers interpreted the same echocardiographic videos and prioritized the regions of interest on myocardium for the differentiation. Based on visualization results, we completed optical flow measurement, myocardial strain, and Doppler/tissue Doppler echocardiography studies to investigate regional myocardial temporal dynamics and diastology. While human readers' visualization predominantly focused on the apex of the heart in TTS patients, the DCNN temporal arm's saliency visualization was attentive on the base of the heart, particularly at the atrioventricular (AV) plane. Compared with STEMI patients, TTS patients consistently showed weaker peak longitudinal displacement (in pixels) in the basal inferoseptal (systolic: 2.15 ± 1.41 vs. 3.10 ± 1.66, P < 0.001; diastolic: 2.36 ± 1.71 vs. 2.97 ± 1.69, P = 0.004) and basal anterolateral (systolic: 2.70 ± 1.96 vs. 3.44 ± 2.13, P = 0.003; diastolic: 2.73 ± 1.70 vs. 3.45 ± 2.20, P = 0.002) segments, and worse longitudinal myocardial strain in the basal inferoseptal (-8.5 ± 3.8% vs. -9.9 ± 4.1%, P = 0.013) and basal anterolateral (-8.6 ± 4.2% vs. -10.4 ± 4.1%, P = 0.006) segments. Meanwhile, TTS patients showed worse diastolic mechanics than STEMI patients (E'/septal: 5.1 ± 1.2 cm/s vs. 6.3 ± 1.5 cm/s, P < 0.001; S'/septal: 5.8 ± 1.3 cm/s vs. 6.8 ± 1.4 cm/s, P < 0.001; E'/lateral: 6.0 ± 1.4 cm/s vs. 7.9 ± 1.6 cm/s, P < 0.001; S'/lateral: 6.3 ± 1.4 cm/s vs. 7.3 ± 1.5 cm/s, P < 0.001; E/E': 15.5 ± 5.6 vs. 12.5 ± 3.5, P < 0.001). Conclusion: The spatiotemporal DCNN saliency visualization helps identify the pattern of myocardial temporal dynamics and navigates the quantification of regional myocardial mechanics. Reduced AV plane displacement in TTS patients likely correlates with impaired diastolic mechanics.

A Parallel Convolutional Network Based on Spiking Neural Systems.

Deep convolutional neural networks have shown advanced performance in accurately segmenting images. In this paper, an SNP-like convolutional neuron structure is introduced, abstracted from the nonlinear mechanism in nonlinear spiking neural P (NSNP) systems. Then, a U-shaped convolutional neural network named SNP-like parallel-convolutional network, or SPC-Net, is constructed for segmentation tasks. The dual-convolution concatenate (DCC) and dual-convolution addition (DCA) network blocks are designed, respectively, in the encoder and decoder stages. The two blocks employ parallel convolution with different kernel sizes to improve feature representation ability and make full use of spatial detail information. Meanwhile, different feature fusion strategies are used to fuse their features to achieve feature complementarity and augmentation. Furthermore, a dual-scale pooling (DSP) module in the bottleneck is designed to improve the feature extraction capability, which can extract multi-scale contextual information and reduce information loss while extracting salient features. The SPC-Net is applied in medical image segmentation tasks and is compared with several recent segmentation methods on the GlaS and CRAG datasets. The proposed SPC-Net achieves 90.77% DICE coefficient, 83.76% IoU score and 83.93% F1 score, 86.33% ObjDice coefficient, 135.60 Obj-Hausdorff distance, respectively. The experimental results show that the proposed model can achieve good segmentation performance.

Integrated Fruit Ripeness Assessment System Based on an Artificial Olfactory Sensor and Deep Learning.

Artificial scent screening systems, inspired by the mammalian olfactory system, hold promise for fruit ripeness detection, but their commercialization is limited by low sensitivity or pattern recognition inaccuracy. This study presents a portable fruit ripeness prediction system based on colorimetric sensing combinatorics and deep convolutional neural networks (DCNN) to accurately identify fruit ripeness. Using the gas chromatography-mass spectrometry (GC-MS) method, the study discerned the distinctive gases emitted by mango, peach, and banana across various ripening stages. The colorimetric sensing combinatorics utilized 25 dyes sensitive to fruit volatile gases, generating a distinct scent fingerprint through cross-reactivity to diverse concentrations and varieties of gases. The unique scent fingerprints can be identified using DCNN. After capturing colorimetric sensor image data, the densely connected convolutional network (DenseNet) was employed, achieving an impressive accuracy rate of 97.39% on the validation set and 82.20% on the test set in assessing fruit ripeness. This fruit ripeness prediction system, coupled with a DCNN, successfully addresses the issues of complex pattern recognition and low identification accuracy. Overall, this innovative tool exhibits high accuracy, non-destructiveness, practical applicability, convenience, and low cost, making it worth considering and developing for fruit ripeness detection.

Artificial intelligence-based framework to identify the abnormalities in the COVID-19 disease and other common respiratory diseases from digital stethoscope data using deep CNN.

The utilization of lung sounds to diagnose lung diseases using respiratory sound features has significantly increased in the past few years. The Digital Stethoscope data has been examined extensively by medical researchers and technical scientists to diagnose the symptoms of respiratory diseases. Artificial intelligence-based approaches are applied in the real universe to distinguish respiratory disease signs from human pulmonary auscultation sounds. The Deep CNN model is implemented with combined multi-feature channels (Modified MFCC, Log Mel, and Soft Mel) to obtain the sound parameters from lung-based Digital Stethoscope data. The model analysis is observed with max-pooling and without max-pool operations using multi-feature channels on respiratory digital stethoscope data. In addition, COVID-19 sound data and enriched data, which are recently acquired data to enhance model performance using a combination of L2 regularization to overcome the risk of overfitting because of less respiratory sound data, are included in the work. The suggested DCNN with Max-Pooling on the improved dataset demonstrates cutting-edge performance employing a multi-feature channels spectrogram. The model has been developed with different convolutional filter sizes (1×12, 1×24, 1×36, 1×48, and 1×60) that helped to test the proposed neural network. According to the experimental findings, the suggested DCNN architecture with a max-pooling function performs better to identify respiratory disease symptoms than DCNN without max-pooling. In order to demonstrate the model's effectiveness in categorization, it is trained and tested with the DCNN model that extract several modalities of respiratory sound data.

Accurate staging of chick embryonic tissues via deep learning of salient features.

Recent work shows that the developmental potential of progenitor cells in the HH10 chick brain changes rapidly, accompanied by subtle changes in morphology. This demands increased temporal resolution for studies of the brain at this stage, necessitating precise and unbiased staging. Here, we investigated whether we could train a deep convolutional neural network to sub-stage HH10 chick brains using a small dataset of 151 expertly labelled images. By augmenting our images with biologically informed transformations and data-driven preprocessing steps, we successfully trained a classifier to sub-stage HH10 brains to 87.1% test accuracy. To determine whether our classifier could be generally applied, we re-trained it using images (269) of randomised control and experimental chick wings, and obtained similarly high test accuracy (86.1%). Saliency analyses revealed that biologically relevant features are used for classification. Our strategy enables training of image classifiers for various applications in developmental biology with limited microscopy data.

Mobile-DenseNet: Detection of building concrete surface cracks using a new fusion technique based on deep learning.

Crack detection is very important during the inspection of building structures to determine whether they are safe. Therefore, to ensure the reliability and longevity of buildings, it is necessary to have experts periodically carry out building inspections. Building inspection has traditionally been conducted using human-based visual inspection methods as well as artificial intelligence methods that have shown great success in computer vision in recent years. In this study, 9 different models (Xception, VGG16, ResNet101, InceptionV3, InceptionResNetV2, MobileNetV2, DenseNet169, NASNetMobile, and EfficientNetB6), which have shown significant success in the field of artificial intelligence, are discussed to detect and classify cracks in building structures. In addition, a new fusion model structure called Mobile-DenseNet has been proposed by making block cutting and adding auxiliary layers to the MobileNetV2 and DenseNet169 model structures. With this proposed model structure, cracks in concrete structures were classified. A dataset consisting of concrete surface images was used to detect and classify cracks occurring in concrete structures, and a 99.87 % success rate was achieved with the proposed Mobile-DenseNet model in classifying cracks occurring on the concrete surface. The proposed model outperformed the traditional pretrained model structures in the study in terms of the number of transactions, density, features, complexity, and success accuracy.

CollectiveNet-AltSpec: A collective concurrent CNN architecture of alternate specifications for EEG media perception and emotion tracing aided by multi-domain feature-augmentation.

Enhancing computability of cerebral recordings and connections made with human/non-human brain have been on track and are expected to propel in our current era. An effective contribution towards said ends is improving accuracy of attempts at discerning intricate phenomena taking place within human brain. Here and in two different capacities of experiments, we attempt to distinguish cerebral perceptions shaped and affective states surfaced during observation of samples of media incorporating distinct audio-visual and emotional contents, through employing electroencephalograph/EEG recorded sessions of two reputable datasets of DEAP and SEED. Here we introduce AltSpec(E3) the inceptive form of CollectiveNet intelligent computational architectures employing collective and concurrent multi-spec analysis to exploit complex patterns in complex data-structures. This processing technique uses a full array of diversification protocols with multifarious parts enabling surgical levels of optimization while integrating a holistic analysis of patterns. Data-structures designed here contain multi-electrode neuroinformatic and neurocognitive features studying emotion reactions and attentive patterns. These spatially and temporally featured 2D/3D constructs of domain-augmented data are eventually AI-processed and outputs are defragmented forming one definitive judgement. The media-perception tracing is arguably first of its kind, at least when implemented on mentioned datasets. Backed by this multi-directional approach and in subject-independent configurations for perception-tracing on 5-media-class basis, mean accuracies of 81.00% and 68.93% were obtained on DEAP and SEED, respectively. We also managed to classify emotions with accuracies of 61.59% and 66.21% in cross-dataset validation followed by 81.47% and 88.12% in cross-subject validation settings trained on DEAP and SEED, consecutively.

Pepper leaf disease recognition based on enhanced lightweight convolutional neural networks.

Pepper leaf disease identification based on convolutional neural networks (CNNs) is one of the interesting research areas. However, most existing CNN-based pepper leaf disease detection models are suboptimal in terms of accuracy and computing performance. In particular, it is challenging to apply CNNs on embedded portable devices due to a large amount of computation and memory consumption for leaf disease recognition in large fields. Therefore, this paper introduces an enhanced lightweight model based on GoogLeNet architecture. The initial step involves compressing the Inception structure to reduce model parameters, leading to a remarkable enhancement in recognition speed. Furthermore, the network incorporates the spatial pyramid pooling structure to seamlessly integrate local and global features. Subsequently, the proposed improved model has been trained on the real dataset of 9183 images, containing 6 types of pepper diseases. The cross-validation results show that the model accuracy is 97.87%, which is 6% higher than that of GoogLeNet based on Inception-V1 and Inception-V3. The memory requirement of the model is only 10.3 MB, which is reduced by 52.31%-86.69%, comparing to GoogLeNet. We have also compared the model with the existing CNN-based models including AlexNet, ResNet-50 and MobileNet-V2. The result shows that the average inference time of the proposed model decreases by 61.49%, 41.78% and 23.81%, respectively. The results show that the proposed enhanced model can significantly improve performance in terms of accuracy and computing efficiency, which has potential to improve productivity in the pepper farming industry.

3D X-ray Histology for the Investigation of Temporal Lobe Epilepsy in a Mouse Model.

The most common form of epilepsy among adults is mesial temporal lobe epilepsy (mTLE), with seizures often originating in the hippocampus due to abnormal electrical activity. The gold standard for the histopathological analysis of mTLE is histology, which is a two-dimensional technique. To fill this gap, we propose complementary three-dimensional (3D) X-ray histology. Herein, we used synchrotron radiation-based phase-contrast microtomography with 1.6 µm-wide voxels for the post mortem visualization of tissue microstructure in an intrahippocampal-kainate mouse model for mTLE. We demonstrated that the 3D X-ray histology of unstained, unsectioned, paraffin-embedded brain hemispheres can identify hippocampal sclerosis through the loss of pyramidal neurons in the first and third regions of the Cornu ammonis as well as granule cell dispersion within the dentate gyrus. Morphology and density changes during epileptogenesis were quantified by segmentations from a deep convolutional neural network. Compared to control mice, the total dentate gyrus volume doubled and the granular layer volume quadrupled 21 days after injecting kainate. Subsequent sectioning of the same mouse brains allowed for benchmarking 3D X-ray histology against well-established histochemical and immunofluorescence stainings. Thus, 3D X-ray histology is a complementary neuroimaging tool to unlock the third dimension for the cellular-resolution histopathological analysis of mTLE.

Few-Shot Learning-Based Light-Weight WDCNN Model for Bearing Fault Diagnosis in Siamese Network.

In this study, bearing fault diagnosis is performed with a small amount of data through few-shot learning. Recently, a fault diagnosis method based on deep learning has achieved promising results. Most studies required numerous training samples for fault diagnosis. However, at manufacturing sites, it is impossible to have enough training samples to represent all fault types under all operating conditions. In addition, most studies consider only accuracy, and models are complex and computationally expensive. Research that only considers accuracy is inefficient since manufacturing sites change rapidly. Therefore, in this study, we propose a few-shot learning model that can effectively learn with small data. In addition, a Depthwise Separable Convolution layer that can effectively reduce parameters is used together. In order to find an efficient model, the optimal hyperparameters were found by adjusting the number of blocks and hyperparameters, and by using a Depthwise Separable Convolution layer for the optimal hyperparameters, it showed higher accuracy and fewer parameters than the existing model.

Reduction of SPECT acquisition time using deep learning: A phantom study.

Single photon emission computed tomography (SPECT) procedures are characterized by long acquisition time to acquire diagnostically acceptable image data. The goal of this investigation was to assess the feasibility of using a deep convolutional neural network (DCNN) to reduce the acquisition time. The DCNN was implemented using the PyTorch and trained using image data from standard SPECT quality phantoms. The under-sampled image dataset is provided to neural network as input, while missing projections were provided as targets. The network is to produce for the output the missing projections. The baseline method of calculating the missing projections as arithmetic means of adjacent ones was introduced. The obtained synthesized projections and reconstructed images were compared to original data and baseline data across several parameters using PyTorch and PyTorch Image Quality code libraries. Results obtained from comparisons of projection and reconstructed image data show the DCNN clearly outperforming the baseline method. However, subsequent analysis revealed the synthesized image data being more comparable to under-sampled than to fully-sampled image data. The results of this investigation imply that neural network can replicate coarser objects better. However, densely sampled clinical image datasets, coarse reconstruction matrices and patient data featuring coarse structures combined with a lack of baseline data generation methods will hamper the ability to analyse the neural network outputs correctly. This study calls for use of phantom image data and introduction of a baseline method in the evaluation of neural network outputs.

Automated Uterine Fibroids Detection in Ultrasound Images Using Deep Convolutional Neural Networks.

Fibroids of the uterus are a common benign tumor affecting women of childbearing age. Uterine fibroids (UF) can be effectively treated with earlier identification and diagnosis. Its automated diagnosis from medical images is an area where deep learning (DL)-based algorithms have demonstrated promising results. In this research, we evaluated state-of-the-art DL architectures VGG16, ResNet50, InceptionV3, and our proposed innovative dual-path deep convolutional neural network (DPCNN) architecture for UF detection tasks. Using preprocessing methods including scaling, normalization, and data augmentation, an ultrasound image dataset from Kaggle is prepared for use. After the images are used to train and validate the DL models, the model performance is evaluated using different measures. When compared to existing DL models, our suggested DPCNN architecture achieved the highest accuracy of 99.8 percent. Findings show that pre-trained deep-learning model performance for UF diagnosis from medical images may significantly improve with the application of fine-tuning strategies. In particular, the InceptionV3 model achieved 90% accuracy, with the ResNet50 model achieving 89% accuracy. It should be noted that the VGG16 model was found to have a lower accuracy level of 85%. Our findings show that DL-based methods can be effectively utilized to facilitate automated UF detection from medical images. Further research in this area holds great potential and could lead to the creation of cutting-edge computer-aided diagnosis systems. To further advance the state-of-the-art in medical imaging analysis, the DL community is invited to investigate these lines of research. Although our proposed innovative DPCNN architecture performed best, fine-tuned versions of pre-trained models like InceptionV3 and ResNet50 also delivered strong results. This work lays the foundation for future studies and has the potential to enhance the precision and suitability with which UF is detected.

Multi-path decoder U-Net: A weakly trained real-time segmentation network for object detection and localization in ultrasound scans.

Detecting and localizing an anatomical structure of interest within the field of view of an ultrasound scan is an essential step in many diagnostic and therapeutic procedures. However, ultrasound scans suffer from high levels of variabilities across sonographers and patients, making it challenging for sonographers to accurately identify and locate these structures without extensive experience. Segmentation-based convolutional neural networks (CNNs) have been proposed as a solution to assist sonographers in this task. Despite their accuracy, these networks require pixel-wise annotations for training; an expensive and labor-intensive operation that requires the expertise of an experienced practitioner to identify the precise outline of the structures of interest. This complicates, delays, and increases the cost of network training and deployment. To address this problem, we propose a multi-path decoder U-Net architecture that is trained on bounding box segmentation maps; not requiring pixel-wise annotations. We show that the network can be trained on small training sets, which is the case in medical imaging datasets; reducing the cost and time needed for deployment and use in clinical settings. The multi-path decoder design allows for better training of deeper layers and earlier attention to the target anatomical structures of interest. This architecture offers up to a 7% relative improvement compared to the U-Net architecture in localization and detection performance, with an increase of only 0.75% in the number of parameters. Its performance is on par with, or slightly better than, the more computationally expensive U-Net++, which has 20% more parameters; making the proposed architecture a more computationally efficient alternative for real-time object detection and localization in ultrasound scans.