Convolutional Neural Network-based Algorithm Research Articles

Implementation of Smart Farm Systems Based on Fog Computing in Artificial Intelligence of Things Environments.

Cloud computing has recently gained widespread attention owing to its use in applications involving the Internet of Things (IoT). However, the transmission of massive volumes of data to a cloud server often results in overhead. Fog computing has emerged as a viable solution to address this issue. This study implements an Artificial Intelligence of Things (AIoT) system based on fog computing on a smart farm. Three experiments are conducted to evaluate the performance of the AIoT system. First, network traffic volumes between systems employing and not employing fog computing are compared. Second, the performance of the communication protocols-hypertext transport protocol (HTTP), message queuing telemetry transport protocol (MQTT), and constrained application protocol (CoAP)-commonly used in IoT applications is assessed. Finally, a convolutional neural network-based algorithm is introduced to determine the maturity level of coffee tree images. Experimental data are collected over ten days from a coffee tree farm in the Republic of Korea. Notably, the fog computing system demonstrates a 26% reduction in the cumulative data volume compared with a non-fog system. MQTT exhibits stable results in terms of the data volume and loss rate. Additionally, the maturity level determination algorithm performed on coffee fruits provides reliable results.

Multimodal Convolutional Neural Network-Based Algorithm for Real-Time Detection and Differentiation of Malignant and Inflammatory Biliary Strictures in Cholangioscopy: a Proof of Concept Study with Video

Deep learning algorithms gained attention for detection (computer-aided detection [CADe]) of biliary tract cancer in digital single-operator cholangioscopy (dSOC). We developed a multimodal convolutional neural network (CNN) for detection (CADe), characterization and discriminating (computer-aided diagnosis [CADx]) between malignant, inflammatory, and normal biliary tissue in raw dSOC videos. In addition, clinical metadata were included in the CNN algorithm to overcome limitations of image-only models. Based on dSOC videos and images of 111 patients (total of 15,158 still frames), a real-time CNN-based algorithm for CADe and CADx was developed and validated. We established an image-only model and metadata injection approach. In addition, frame-wise and case-based predictions on complete dSOC video sequences were validated. Model embeddings were visualized, and class activation maps highlighted relevant image regions. The concatenation-based CADx approach achieved a per-frame area under the receiver-operating characteristic curve of .871, sensitivity of .809 (95% CI, .784-.832), specificity of .773 (95% CI, .761-.785), positive predictive value of .450 (95% CI, .423-.467), and negative predictive value of .946 (95% CI, .940-.954) with respect to malignancy on 5715 test frames from complete videos of 20 patients. For case-based diagnosis using average prediction scores, 6 of 8 malignant cases and all 12 benign cases were identified correctly. Our algorithm distinguishes malignant and inflammatory bile duct lesions in dSOC videos, indicating the potential of CNN-based diagnostic support systems for both CADe and CADx. The integration of non-image data can improve CNN-based support systems, targeting current challenges in the assessment of biliary strictures.

Achieving Image Encryption Quantum Dot-Functionalized Encryption Camera with Designed Films.

The risk of information leaks increases as images become a crucial medium for information sharing. There is a great need to further develop the versatility of image encryption technology to protect confidential and sensitive information. Herein, using high spatial redundancy (strong correlation of neighboring pixels) of the image and the in situ encryption function of a quantum dot functionalized encryption camera, in situ image encryption is achieved by designing quantum dot films (size, color, and full width at half maximum) to modify the correlation and reduce spatial redundancy of the captured image during encryption processing. The correlation coefficients of simulated encrypted image closely apporach to 0. High-quality decrypted images are achieved with a PSNR of more than 35dB by a convolutional neural network-based algorithm that meets the resolution requirements of human visual perception. Compared with the traditional image encryption algorithms, chaotic image encryption algorithms and neural network-based encryption algorithms described previously, it provides a universal, efficient and effective in situ image encryption method.

An overview of hand gesture recognition based on computer vision

Hand gesture recognition emerges as one of the foremost sectors which has gone through several developments within pattern recognition. Numerous studies and research endeavors have explored methodologies grounded in computer vision within this domain. Despite extensive research endeavors, there is still a need for a more thorough evaluation of the efficiency of various methods in different environments along with the challenges encountered during the application of these methods. The focal point of this paper is the comparison of different research in the domain of vision-based hand gesture recognition. The objective is to find out the most prominent methods by reviewing efficiency. Concurrently, the paper delves into presenting potential solutions for challenges faced in different research. A comparative analysis particularly centered around traditional methods and convolutional neural networks like random forest, long short-term memory (LSTM), heatmap, and you only look once (YOLO). considering their efficacy. Where convolutional neural network-based algorithms performed best for recognizing the gestures and gave effective solutions for the challenges faced by the researchers. In essence, the findings of this review paper aim to contribute to future implementations and the discovery of more efficient approaches in the gesture recognition sector.

Practical Application and Case Analysis of Computer Image Vision Technology in Music Education and Teaching

Exploring how to utilize images to enrich music teaching content and provide a more visually impactful learning experience is an important topic. Therefore, this paper introduces a convolutional neural network-based algorithm for extracting audio features to construct a music visualization model. By identifying features such as note pitches, it enhances pitch recognition and integrates with CNN algorithms for audio information visualization. Experimental results demonstrate an accuracy rate exceeding 97%, showcasing the significant advantage of this method in visualizing audio information in music multimedia classrooms. It provides technical support for bringing a new visual experience to music education.

Improving Earth-like planet detection in radial velocity using deep learning

Context. Many novel methods have been proposed to mitigate stellar activity for exoplanet detection as the presence of stellar activity in radial velocity (RV) measurements is the current major limitation. Unlike traditional methods that model stellar activity in the RV domain, more methods are moving in the direction of disentangling stellar activity at the spectral level. As deep neural networks have already been proven to be one of the most effective tools in data mining, in this work, we explore their potential in the context of Earth-like planet detection in RV measurements. Aims. The goal of this paper is to present a novel convolutional neural network-based algorithm that efficiently models stellar activity signals at the spectral level, enhancing the detection of Earth-like planets. Methods. Based on the idea that the presence of planets can only produce a Doppler shift at the spectral level while the presence of stellar activity can introduce a variation in the profile of spectral lines (asymmetry and depth change), we trained a convolutional neural network to build the correlation between the change in the spectral line profile and the corresponding RV, full width at half maximum (FWHM) and bisector span (BIS) values derived from the classical cross-correlation function. Results. This algorithm has been tested on three intensively observed stars: Alpha Centauri B (HD 128621), Tau ceti (HD 10700), and the Sun. By injecting simulated planetary signals at the spectral level, we demonstrate that our machine learning algorithm can achieve, for HD 128621 and HD 10700, a detection threshold of 0.5 m s−1 in semi-amplitude for planets with periods ranging from 10 to 300 days. This threshold would correspond to the detection of a ~4 M⊕ in the habitable zone of those stars. On the HARPS-N solar dataset, thanks to significantly more data, our algorithm is even more efficient at mitigating stellar activity signals and can reach a threshold of 0.2 m s−1, which would correspond to a 2.2 M⊕ planet on the orbit of the Earth. Conclusions. To the best of our knowledge, it is the first time that such low detection thresholds are reported for the Sun, but also for other stars, and therefore this highlights the efficiency of our convolutional neural network-based algorithm at mitigating stellar activity in RV measurements.

A Deep Learning Approach for Improving Two-Photon Vascular Imaging Speeds.

A potential method for tracking neurovascular disease progression over time in preclinical models is multiphoton fluorescence microscopy (MPM), which can image cerebral vasculature with capillary-level resolution. However, obtaining high-quality, three-dimensional images with traditional point scanning MPM is time-consuming and limits sample sizes for chronic studies. Here, we present a convolutional neural network-based (PSSR Res-U-Net architecture) algorithm for fast upscaling of low-resolution or sparsely sampled images and combine it with a segmentation-less vectorization process for 3D reconstruction and statistical analysis of vascular network structure. In doing so, we also demonstrate that the use of semi-synthetic training data can replace the expensive and arduous process of acquiring low- and high-resolution training pairs without compromising vectorization outcomes, and thus open the possibility of utilizing such approaches for other MPM tasks where collecting training data is challenging. We applied our approach to images with large fields of view from a mouse model and show that our method generalizes across imaging depths, disease states and other differences in neurovasculature. Our pretrained models and lightweight architecture can be used to reduce MPM imaging time by up to fourfold without any changes in underlying hardware, thereby enabling deployability across a range of settings.

Retracted: Convolutional Neural Network-Based Algorithm for Predicting the Gross Marine Product in Fujian Province

Retracted: Convolutional Neural Network-Based Algorithm for Predicting the Gross Marine Product in Fujian Province

AI-Supported Autonomous Uterus Reconstructions: First Application in MRI Using 3D SPACE with Iterative Denoising

T2-weighted imaging in at least two orthogonal planes is recommended for assessment of the uterus. To determine whether a convolutional neural network-based algorithm could be used for the re-constructions of uterus axes derived from a 3D SPACE with iterative denoising. 50 patients aged 18-81 (mean: 42) years who underwent an MRI examination of the uterus participated voluntarily in this prospective study after informed consent. In addition to a standard MRI pelvis protocol, a 3D SPACE research application sequence was acquired in sagittal orientation. Reconstructions for both the cervix and the cavum in the short and long axes were performed by a research trainee (T), an experienced radiologist (E), and the prototype software (P). In thenext step, the reconstructions were evaluated anonymously by two experienced readers according to 5-point-Likert-Scales. In addition, the length of the cervical canal, the length of the cavum and the distance between the tube angles were measured on all reconstructions. Interobserver agreement was assessed for all ratings. For all axes, significant differences were found between the scores of the reconstructions by research T, E and P. P received higher scores and was preferred significantly more often with the exception of the comparison of the reconstruction Cervix short of E (Cervix short: P vs. T: p=0.02; P vs. E: p=0.26; Cervix long: P vs. T: p=0.01; P vs. E: p<0.01; Cavum short: P vs. T: p=0.01; P vs. E: p=0.02; Cavum long: P vs. T: p<0.01; P vs. E: p<0.01). Regarding the measured diameters, (length of cervical canal/cavum/distance between tube angles) significantly larger diameters were recorded for P compared to E and T (Cervix long (mm): T: 25.43; E: 25.65; P: 26.65; Cavum short (mm): T: 26.24; E: 25.04; P: 27.33; Cavum long (mm): T: 31.98; E: 32.91; P: 34.41; P vs. T: p<0.01); P vs. E: p=0.04). Moderate to substantial agreement was found between Reader 1 and Reader 2 (range: 0.39-0.67). P was able to reconstruct the axes at least as well as or better than E and T. P could thereby lead to workflow facilitation and enable more efficient reporting of uterine MRI.

Emotion Recognition in Individuals with Down Syndrome: A Convolutional Neural Network-Based Algorithm Proposal

This research introduces an algorithm that automatically detects five primary emotions in individuals with Down syndrome: happiness, anger, sadness, surprise, and neutrality. The study was conducted in a specialized institution dedicated to caring for individuals with Down syndrome, which allowed for collecting samples in uncontrolled environments and capturing spontaneous emotions. Collecting samples through facial images strictly followed a protocol approved by certified Ethics Committees in Ecuador and Colombia. The proposed system consists of three convolutional neural networks (CNNs). The first network analyzes facial microexpressions by assessing the intensity of action units associated with each emotion. The second network utilizes transfer learning based on the mini-Xception architecture, using the Dataset-DS, comprising images collected from individuals with Down syndrome as the validation dataset. Finally, these two networks are combined in a CNN network to enhance accuracy. The final CNN processes the information, resulting in an accuracy of 85.30% in emotion recognition. In addition, the algorithm was optimized by tuning specific hyperparameters of the network, leading to a 91.48% accuracy in emotion recognition accuracy, specifically for people with Down syndrome.

Comparative Evaluation of Breast Ductal Carcinoma Grading: A Deep-Learning Model and General Pathologists' Assessment Approach.

Breast cancer is the most prevalent neoplasia among women, with early and accurate diagnosis critical for effective treatment. In clinical practice, however, the subjective nature of histological grading of infiltrating ductal adenocarcinoma of the breast (DAC-NOS) often leads to inconsistencies among pathologists, posing a significant challenge to achieving optimal patient outcomes. Our study aimed to address this reproducibility problem by leveraging artificial intelligence (AI). We trained a deep-learning model using a convolutional neural network-based algorithm (CNN-bA) on 100 whole slide images (WSIs) of DAC-NOS from the Cancer Genome Atlas Breast Invasive Carcinoma (TCGA-BRCA) dataset. Our model demonstrated high precision, sensitivity, and F1 score across different grading components in about 17.5 h with 19,000 iterations. However, the agreement between the model's grading and that of general pathologists varied, showing the highest agreement for the mitotic count score. These findings suggest that AI has the potential to enhance the accuracy and reproducibility of breast cancer grading, warranting further refinement and validation of this approach.

An audio-based vehicle classifier using convolutional neural network

Audio-based event and scene classification are getting more attention in recent years. Many examples of environmental noise detection, vehicle classification, and soundscape analysis are developed using state of art deep learning techniques. The major noise source in urban and rural areas is traffic noise. Environmental noise parameters for urban and rural small roads have not been investigated due to some practical reasons. The purpose of this study is to develop an audio-based traffic classifier for rural and urban small roads which have limited or no traffic flow data to supply values for noise mapping and other noise metrics. To develop an audio-based vehicle classifier a convolutional neural network-based algorithm was proposed using Mel spectrogram of audio signals as an input feature. Different variations of the network were generated by changing the parameters of the convolutional layers and the length of the network. Filter size, number of filters were tested with a dataset prepared with various real-life traffic records and audio extracts from traffic videos. The precision of the networks was evaluated with the common performance metrics. Further assessments were conducted with longer audio files and predictions of the system compared with actual traffic flow. The results showed that convolutional neural networks can be used to classify road traffic noise sources and perform outstandingly for single or double-lane roads.

Research on Volume Measurement of Logs Based on Embedded Application

Spurred by the worldwide concern for forest protection and the increased log sales, most countries have standardized log volume calculations to avoid excessive timber and protect buyers. However, log volume is currently manually measured, suffering from high labor costs, low measurement progress, and imposing significant measurement errors. Thus, automatically obtaining the volumetric data of logs is a convenient and quick solution. Therefore, this work proposes a Mask Region Convolutional Neural Network-based (R-CNN) algorithm for Logs volume measurement, named Wood Mask (WM) R-CNN. Specifically, we employ the Res2Net structure as the backbone to obtain receptive fields that exceed the input feature size, thus improving our model’s multi-scale information fusion ability. Additionally, WM R-CNN relies on the Path Aggregation Feature Pyramid Network’s (PAFPN’s) path enhancement structure, shortening the low-level feature map’s propagation path and improving the wood contour segmentation accuracy. Extensive experiments on the Vehicle-mounted Dense Logs (VMDL) dataset demonstrate that WM R-CNN affords a highly appealing segmentation accuracy for small, medium, and large wood, improving the corresponding mAP indicators against current methods by 2.0%, 1.2%, and 4.4%, respectively. Furthermore, a quantitative method based on TensorRT compresses the proposed model to deploy the WM R-CNN to mobile embedded devices. However, to compensate for the quantization loss, we introduce the expansion convolution operation method to manipulate the mask map and control the volume calculation error of all logs on a vehicle within 1%. The experiments reveal that the proposed method offers an appealing performance, verifying the algorithm’s effectiveness and implementation ability on mobile terminals.

External clinical validation of automated software to identify structural abnormalities and microhaemorrhages in nailfold videocapillaroscopy images.

Automated systems to analyse nailfold videocapillaroscopy (NVC) images are needed to promptly and comprehensively characterise patients with systemic sclerosis (SSc) or Raynaud's phenomenon (RP). We previously developed, and validated in-house, a deep convolutional neural network-based algorithm to classify NVC-captured images according to the presence/absence of structural abnormalities and/or microhaemorrhages. We present its external clinical validation. A total of 1,164 NVC images of RP patients were annotated by 5 trained capillaroscopists according to the following categories: normal capillary; dilation; giant capillary; abnormal shape; tortuosity; microhaemorrhage. The images were also presented to the algorithm. Matches and discrepancies between algorithm predictions and those annotations obtained by consensus of ≥3 or ≥4 interobservers were analysed. Consensus among ≥3 capillaroscopists was achieved in 86.9% of images, 75.8% of which were correctly predicted by the algorithm. Consensus among ≥4 experts occurred in 52.0% of cases, in which 87.1% of the algorithm's results matched with those of the expert panel. The algorithm's positive predictive value was >80% for microhaemorrhages and unaltered, giant or abnormal capillaries. Sensitivity was >75% for dilations and tortuosities. Negative predictive value and specificity were >89% for all categories. This external clinical validation suggests that this algorithm is useful to assist in the diagnosis and follow-up of SSc or RP patients in a timely manner. It may also be helpful in the management of patients with any pathology presenting with microvascular changes, as the algorithm has been designed to also be useful for research aiming at extending the usage of nailfold capillaroscopy to more conditions.

A Semantic Data-Based Distributed Computing Framework to Accelerate Digital Twin Services for Large-Scale Disasters.

As natural disasters become extensive, due to various environmental problems, such as the global warming, it is difficult for the disaster management systems to rapidly provide disaster prediction services, due to complex natural phenomena. Digital twins can effectively provide the services using high-fidelity disaster models and real-time observational data with distributed computing schemes. However, the previous schemes take little account of the correlations between environmental data of disasters, such as landscapes and weather. This causes inaccurate computing load predictions resulting in unbalanced load partitioning, which increases the prediction service times of the disaster management agencies. In this paper, we propose a novel distributed computing framework to accelerate the prediction services through semantic analyses of correlations between the environmental data. The framework combines the data into disaster semantic data to represent the initial disaster states, such as the sizes of wildfire burn scars and fuel models. With the semantic data, the framework predicts computing loads using the convolutional neural network-based algorithm, partitions the simulation model into balanced sub-models, and allocates the sub-models into distributed computing nodes. As a result, the proposal shows up to 38.5% of the prediction time decreases, compared to the previous schemes.

Development of Deep Convolutional Neural Network-Based Algorithm for Muscle Biopsy Diagnosis

To diagnose muscle disease, histopathologic evaluation of muscle biopsy is essential. In addition, since myositis has a well-established treatment, an accurate diagnosis is required. However, distinguishing myositis from other muscle diseases is challenging for pathologists. Thus, artificial intelligence is expected to improve medical productivity. Therefore, we developed an algorithm based on deep convolutional neural networks to make the algorithm for muscle biopsy diagnosis. We used 1,400 hematoxylin-and-eosin-stained pathology slides for training and testing. Our trained algorithm achieved better sensitivity and specificity than the diagnoses made by physicians.

Development and Validation of an Artificial Intelligence Model for Small Bowel Capsule Endoscopy Video Review

Reading small bowel capsule endoscopy (SBCE) videos is a tedious task for clinicians, and a new method should be applied to solve the situation. To develop and evaluate the performance of a convolutional neural network algorithm for SBCE video review in real-life clinical care. In this multicenter, retrospective diagnostic study, a deep learning neural network (SmartScan) was trained and validated for the SBCE video review. A total of 2927 SBCE examinations from 29 medical centers were used to train SmartScan to detect 17 types of CE structured terminology (CEST) findings from January 1, 2019, to June 30, 2020. SmartScan was later validated with conventional reading (CR) and SmartScan-assisted reading (SSAR) in 2898 SBCE examinations collected from 22 medical centers. Data analysis was performed from January 25 to December 31, 2021. An artificial intelligence-based tool for interpreting clinical images of SBCE. The detection rate and efficiency of CEST findings detected by SSAR and CR were compared. A total of 5825 SBCE examinations were retrospectively collected; 2898 examinations (1765 male participants [60.9%]; mean [SD] age, 49.8 [15.5] years) were included in the validation phase. From a total of 6084 CEST-classified SB findings, SSAR detected 5834 findings (95.9%; 95% CI, 95.4%-96.4%), significantly higher than CR, which detected 4630 findings (76.1%; 95% CI, 75.0%-77.2%). SmartScan-assisted reading achieved a higher per-patient detection rate (79.3% [2298 of 2898]) for CEST findings compared with CR (70.7% [2048 of 2298]; 95% CI, 69.0%-72.3%). With SSAR, the mean (SD) number of images (per SBCE video) requiring review was reduced to 779.2 (337.2) compared with 27 910.8 (12 882.9) with CR, for a mean (SD) reduction rate of 96.1% (4.3%). The mean (SD) reading time with SSAR was shortened to 5.4 (1.5) minutes compared with CR (51.4 [11.6] minutes), for a mean (SD) reduction rate of 89.3% (3.1%). This study suggests that a convolutional neural network-based algorithm is associated with an increased detection rate of SBCE findings and reduced SBCE video reading time.

AI-based automatic segmentation of craniomaxillofacial anatomy from CBCT scans for automatic detection of pharyngeal airway evaluations in OSA patients

This study aims to generate and also validate an automatic detection algorithm for pharyngeal airway on CBCT data using an AI software (Diagnocat) which will procure a measurement method. The second aim is to validate the newly developed artificial intelligence system in comparison to commercially available software for 3D CBCT evaluation. A Convolutional Neural Network-based machine learning algorithm was used for the segmentation of the pharyngeal airways in OSA and non-OSA patients. Radiologists used semi-automatic software to manually determine the airway and their measurements were compared with the AI. OSA patients were classified as minimal, mild, moderate, and severe groups, and the mean airway volumes of the groups were compared. The narrowest points of the airway (mm), the field of the airway (mm2), and volume of the airway (cc) of both OSA and non-OSA patients were also compared. There was no statistically significant difference between the manual technique and Diagnocat measurements in all groups (p > 0.05). Inter-class correlation coefficients were 0.954 for manual and automatic segmentation, 0.956 for Diagnocat and automatic segmentation, 0.972 for Diagnocat and manual segmentation. Although there was no statistically significant difference in total airway volume measurements between the manual measurements, automatic measurements, and DC measurements in non-OSA and OSA patients, we evaluated the output images to understand why the mean value for the total airway was higher in DC measurement. It was seen that the DC algorithm also measures the epiglottis volume and the posterior nasal aperture volume due to the low soft-tissue contrast in CBCT images and that leads to higher values in airway volume measurement.

Testing the reliability of monocular obstacle detection methods in a simulated 3D factory environment

Automated driving in public traffic still faces many technical and legal challenges. However, automating vehicles at low speeds in controlled industrial environments is already achievable today. A reliable obstacle detection is mandatory to prevent accidents. Recent advances in convolutional neural network-based algorithms hypothetically allow the replacement of distance measuring laser scanners with common monocameras. In this paper, we present a photorealistic 3D simulated factory environment for testing vision-based obstacle detecting algorithms preceding field tests on the safety–critical system. We further test two obstacle detection methods employing state-of-the-art semantic segmentation and depth estimation in a range of challenging test scenarios. Both models performed well under common factory settings. Some edge cases, however, lead to vehicle crashes.

An Efficient Automatic Fruit-360 Image Identification and Recognition Using a Novel Modified Cascaded-ANFIS Algorithm.

Automated fruit identification is always challenging due to its complex nature. Usually, the fruit types and sub-types are location-dependent; thus, manual fruit categorization is also still a challenging problem. Literature showcases several recent studies incorporating the Convolutional Neural Network-based algorithms (VGG16, Inception V3, MobileNet, and ResNet18) to classify the Fruit-360 dataset. However, none of them are comprehensive and have not been utilized for the total 131 fruit classes. In addition, the computational efficiency was not the best in these models. A novel, robust but comprehensive study is presented here in identifying and predicting the whole Fruit-360 dataset, including 131 fruit classes with 90,483 sample images. An algorithm based on the Cascaded Adaptive Network-based Fuzzy Inference System (Cascaded-ANFIS) was effectively utilized to achieve the research gap. Color Structure, Region Shape, Edge Histogram, Column Layout, Gray-Level Co-Occurrence Matrix, Scale-Invariant Feature Transform, Speeded Up Robust Features, Histogram of Oriented Gradients, and Oriented FAST and rotated BRIEF features are used in this study as the features descriptors in identifying fruit images. The algorithm was validated using two methods: iterations and confusion matrix. The results showcase that the proposed method gives a relative accuracy of 98.36%. The Fruit-360 dataset is unbalanced; therefore, the weighted precision, recall, and FScore were calculated as 0.9843, 0.9841, and 0.9840, respectively. In addition, the developed system was tested and compared against the literature-found state-of-the-art algorithms for the purpose. Comparison studies present the acceptability of the newly developed algorithm handling the whole Fruit-360 dataset and achieving high computational efficiency.