Development Of Deep Learning Algorithms Research Articles

Abstract PO4-20-01: A PROSPECTIVE PILOT STUDY OF MAMMOGRAPHIC PREDICTION OF WOMEN’S RISK (M-POWER) IN NIGERIA

Abstract Background There have been no sustained efforts at stemming the rising incidence and mortality from breast cancer in sub-Saharan African countries like Nigeria. Mammography is the gold standard for breast cancer screening and is associated with improved survival particularly among women 50 to 69. Reports from various parts of Nigeria show very low mammography screening uptake. Causes include lack of geographical access, financial issues (less than 5% of the population is covered by the National Health Insurance Scheme), lack of awareness and the increasing expertise gap from the continued brain drain, despite the already low number of radiologists in the country. Given these obstacles, our long-term goal is to devise a rational basis for identifying mammographic risk categories so that this resource can be deployed cost-effectively. We propose to carry out a 5-year prospective mammographic study in Nigerian women and to use machine learning to develop a risk prediction algorithm, so that following a single screening mammogram, the appropriate frequency of subsequent mammograms can be determined. Our immediate goal is to examine the feasibility of performing such a study. To this end, we have initiated a cross-sectional pilot study that will recruit up to 2000 women in multiple urban centers in Nigeria. Objective: To explore feasibility of creating a digital mammography image database representative of the radiological features of the breast of Nigerian women. Objective: To explore feasibility of creating a database that can be used to inform a predictive mammographic risk model for breast cancer and identify the appropriate population that will benefit from breast cancer screening in Nigeria. Design This is a prospective, open-label, cross-sectional pilot study in 2000 eligible women in Nigeria and will include a single screening mammography examination. The mammography images will be read by the study radiologist. The Pilot phase will determine the feasibility and aid in the design of the of the Extension phase in which up to 20,000 participants will undergo yearly mammography examination for 5 years. Incident cases of breast cancer during the study period will permit the development of deep learning algorithms for predicting the risk of breast cancer in this population. Main Eligibility Criteria: Inclusion Criteria All consecutive women aged ≥40 years and < 65 years presenting for screening mammography at the participating institutions will be eligible. Signed study-specific written informed consent. Exclusion Criteria Complaint of a focal dominant lump or a bloody or clear nipple discharge. History of breast implants. Any woman who is pregnant or has reason to believe that she might be pregnant. Statistical Methods Analyses of continuous data will be summarized using descriptive statistics where the following parameters will be reported: Number of observations, Number of missing observations, Mean, Median, Standard deviation (SD), Minimum (Min), Maximum (Max) Categorical data will be presented with absolute and relative frequency (n and %). Present Accrual and Target accrual Two hundred ninety-six women have been enrolled and screening digital mammographic images using GE Digital mammography equipment have been obtained on all participants at 2 study sites in 2 cities in Southern Nigeria. The target accrual is 2000 participants over 12 months. Conclusion Our preliminary experience to date suggests the feasibility of proceeding with a larger prospective study to define mammographic risk groups in Nigerian women which will permit the rational allocation of resources for breast cancer prevention activities in this underserved population. Contact Information: Dr Toyin Shonukan at tshonukan@yemanjacancercare.org Citation Format: Oluwatoyin Shonukan, Hyelakumi Ibrahim, Esther Udoh, Wahab Egbeolu, Kanyinsola Oyeyinka, Jonas Haggstrom. A PROSPECTIVE PILOT STUDY OF MAMMOGRAPHIC PREDICTION OF WOMEN’S RISK (M-POWER) IN NIGERIA [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO4-20-01.

An annotated street view image dataset for automated road damage detection

Road damage is a great threat to the service life and safety of roads, and the early detection of pavement damage can facilitate maintenance and repair. Street view images serve as a new solution for the monitoring of pavement damage due to their wide coverage and regular updates. In this study, a road pavement damage dataset, the Street View Image Dataset for Automated Road Damage Detection (SVRDD), was developed using 8000 street view images acquired from Baidu Maps. Based on these images, over 20,000 damage instances were visually recognized and annotated. These instances were distributed in five administrative districts of Beijing City. Ten well-established object detection algorithms were trained and assessed using the SVRDD dataset. The results have demonstrated the performances of these algorithms in the detection of pavement damages. To the best of our knowledge, SVRDD is the first public dataset based on street view images for pavement damages detection. It can provide reliable data support for future development of deep learning algorithms based on street view images.

A medical image segmentation method for rectal tumors based on multi-scale feature retention and multiple attention mechanisms.

With the continuous development of deep learning algorithms in the field of medical images, models for medical image processing based on convolutional neural networks have made great progress. Since medical images of rectal tumors are characterized by specific morphological features and complex edges that differ from natural images, achieving good segmentation results often requires a higher level of enrichment through the utilization of semantic features. The efficiency of feature extraction and utilization has been improved to some extent through enhanced hardware arithmetic and deeper networks in most models. However, problems still exist with detail loss and difficulty in feature extraction, arising from the extraction of high-level semantic features in deep networks. In this work, a novel medical image segmentation model has been proposed for Magnetic Resonance Imaging (MRI) image segmentation of rectal tumors. The model constructs a backbone architecture based on the idea of jump-connected feature fusion and solves the problems of detail feature loss and low segmentation accuracy using three novel modules: Multi-scale Feature Retention (MFR), Multi-branch Cross-channel Attention (MCA), and Coordinate Attention (CA). Compared with existing methods, our proposed model is able to segment the tumor region more effectively, achieving 97.4% and 94.9% in Dice and mIoU metrics, respectively, exhibiting excellent segmentation performance and computational speed. Our proposed model has improved the accuracy of both lesion region and tumor edge segmentation. In particular, the determination of the lesion region can help doctors identify the tumor location in clinical diagnosis, and the accurate segmentation of the tumor edge can assist doctors in judging the necessity and feasibility of surgery.

Credit Card Fraud Detection Using State-of-the-Art Machine Learning and Deep Learning Algorithms

People can use credit cards for online transactions as it provides an efficient and easy-to-use facility. With the increase in usage of credit cards, the capacity of credit card misuse has also enhanced. Credit card frauds cause significant financial losses for both credit card holders and financial companies. In this research study, the main aim is to detect such frauds, including the accessibility of public data, high-class imbalance data, the changes in fraud nature, and high rates of false alarm. The relevant literature presents many machines learning based approaches for credit card detection, such as Extreme Learning Method, Decision Tree, Random Forest, Support Vector Machine, Logistic Regression and XG Boost. However, due to low accuracy, there is still a need to apply state of the art deep learning algorithms to reduce fraud losses. The main focus has been to apply the recent development of deep learning algorithms for this purpose. Comparative analysis of both machine learning and deep learning algorithms was performed to find efficient outcomes. The detailed empirical analysis is carried out using the European card benchmark dataset for fraud detection. A machine learning algorithm was first applied to the dataset, which improved the accuracy of detection of the frauds to some extent. Later, three architectures based on a convolutional neural network are applied to improve fraud detection performance. Further addition of layers further increased the accuracy of detection. A comprehensive empirical analysis has been carried out by applying variations in the number of hidden layers, epochs and applying the latest models. The evaluation of research work shows the improved results achieved, such as accuracy, f1-score, precision and AUC Curves having optimized values of 99.9%,85.71%,93%, and 98%, respectively. The proposed model outperforms the state-of-the-art machine learning and deep learning algorithms for credit card detection problems. In addition, we have performed experiments by balancing the data and applying deep learning algorithms to minimize the false negative rate. The proposed approaches can be implemented effectively for the real-world detection of credit card fraud

MHCII-peptide presentation: an assessment of the state-of-the-art prediction methods.

Major histocompatibility complex Class II (MHCII) proteins initiate and regulate immune responses by presentation of antigenic peptides to CD4+ T-cells and self-restriction. The interactions between MHCII and peptides determine the specificity of the immune response and are crucial in immunotherapy and cancer vaccine design. With the ever-increasing amount of MHCII-peptide binding data available, many computational approaches have been developed for MHCII-peptide interaction prediction over the last decade. There is thus an urgent need to provide an up-to-date overview and assessment of these newly developed computational methods. To benchmark the prediction performance of these methods, we constructed an independent dataset containing binding and non-binding peptides to 20 human MHCII protein allotypes from the Immune Epitope Database, covering DP, DR and DQ alleles. After collecting 11 known predictors up to January 2022, we evaluated those available through a webserver or standalone packages on this independent dataset. The benchmarking results show that MixMHC2pred and NetMHCIIpan-4.1 achieve the best performance among all predictors. In general, newly developed methods perform better than older ones due to the rapid expansion of data on which they are trained and the development of deep learning algorithms. Our manuscript not only draws a full picture of the state-of-art of MHCII-peptide binding prediction, but also guides researchers in the choice among the different predictors. More importantly, it will inspire biomedical researchers in both academia and industry for the future developments in this field.

Artificial Intelligence for 3D Reconstruction from 2D Panoramic X-rays to Assess Maxillary Impacted Canines.

The objective of this study was to explore the feasibility of current 3D reconstruction in assessing the position of maxillary impacted canines from 2D panoramic X-rays. A dataset was created using pre-treatment CBCT data from a total of 123 patients, comprising 74 patients with impacted canines and 49 patients without impacted canines. From all 74 subjects, we generated a dataset containing paired 2D panoramic X-rays and pseudo-3D images. This pseudo-3D image contained information about the location of the impacted canine in the buccal/lingual, mesial/distal, and apical/coronal positions. These data were utilized to train a deep-learning reconstruction algorithm, a generative AI. The location of the crown of the maxillary impacted canine was determined based on the output of the algorithm. The reconstruction was evaluated using the structure similarity index measure (SSIM) as a metric to indicate the quality of the reconstruction. The prediction of the impacted canine's location was assessed in both the mesiodistal and buccolingual directions. The reconstruction algorithm predicts the position of the impacted canine in the buccal, middle, or lingual position with 41% accuracy, while the mesial and distal positions are predicted with 55% accuracy. The mean SSIM for the output is 0.71, with a range of 0.63 to 0.84. Our study represents the first application of AI reconstruction output for multidisciplinary care involving orthodontists, periodontists, and maxillofacial surgeons in diagnosing and treating maxillary impacted canines. Further development of deep-learning algorithms is necessary to enhance the robustness of dental reconstruction applications.

Forensic Dental Age Estimation Using Modified Deep Learning Neural Network

Dental age is one of the most reliable methods to identify an individual’s age. By using dental panoramic radiography (DPR) images, physicians and pathologists in forensic sciences try to establish the chronological age of individuals with no valid legal records or registered patients. The current methods in practice demand intensive labor, time, and qualified experts. The development of deep learning algorithms in the field of medical image processing has improved the sensitivity of predicting truth values while reducing the processing speed of imaging time. This study proposed an automated approach to estimate the forensic ages of individuals ranging in age from 8 to 68 using 1332 DPR images. Initially, experimental analyses were performed with the transfer learning-based models, including InceptionV3, DenseNet201, EfficientNetB4, MobileNetV2, VGG16, and ResNet50V2; and accordingly, the best-performing model, InceptionV3, was modified, and a new neural network model was developed. Reducing the number of the parameters already available in the developed model architecture resulted in a faster and more accurate dental age estimation. The performance metrics of the results attained were as follows: mean absolute error (MAE) was 3.13, root mean square error (RMSE) was 4.77, and correlation coefficient R2 was 87%. It is conceivable to propose the new model as potentially dependable and practical ancillary equipment in forensic sciences and dental medicine.

Data-driven framework for analyze the visual quality of streets: A systematic measurement of street quality through street view images utilizing deep learning algorithms

The study developed a framework that utilizes deep learning algorithms to analyse the visual quality of streets (VQOS) through street view images (SVIs), while overcoming the constraints observed in existing practices. The study consisted of four main stages, including literature reviews and expert discussions, development of deep learning algorithms, testing of developed algorithms, and validation. The study developed both convolutional neural network (CNN) and feed forward neural network (FFNN) algorithms, using 2684 street view images extracted from Google Street View images and Cityscape datasets, rated by an expert panel and the general public. The proposed framework comprises stages such as the collection and rating of street view images, image processing, developing and training the CNN, and testing and validating stages. The proposed framework achieved 90.51% internal validation accuracy using the accuracy metric in Keras and 86.7% external validation accuracy, with an accepted level of kappa accuracy of 80%. Urban planners, designers, architects, and landscape architects can use this framework as a tool for quantitatively measuring and mapping the visual quality of streets and assessing the impact of proposed developments and guidelines on the visual quality of streets. This proposed framework will enhance the effectiveness of their new proposals and designs.

PERFORMANCE ASSESSMENT OF OBJECT DETECTION FROM MULTI SATELLITES AND AERIAL IMAGES

Abstract. Object detection in remote sensing imagery plays an important role in many applications, such as tracking and change detection. With the development of deep learning algorithms and advancement in hardware systems, improved accuracies have been achieved in the detection of various objects from remote sensing images. However, object detection across heterogeneous remote sensing imagery remains an important issue, particularly for satellite and aerial imagery. The colour variation for the same ground objects, variable resolutions, different platform heights, the parallax effect, and image distortion brought on by diverse shooting angles are the biggest hurdles in satellite-aerial detection applications. The research aims to obtain successful model for detecting aircrafts from satellite and aerial images and reduce cost and the gap of revisit time between sensors. The networks were tested using aerial, GF-2, Jilin-1 (JL-1) and Pleiades satellites test sets after being trained individually using the RGB high-resolution aerial set and panchromatic low-resolution GF-2 satellite set to validate the efficiency of the trained models. Also, the aerial-trained model and GF-2 satellite-trained model as dedicated models were compared with each other, and model trained by all dataset for Object Detection in Aerial Images (DOTA). It is observed that the anchor sizes and augmentation methods can enhance the performance of detection models. k-means algorithm and data augmentation were applied to produce better anchor box selection and avoid overfitting, atmospheric conditions problems, respectively. The accuracy assessment results demonstrate that the aerial-trained model outperforms the GF-2 satellite-trained model. In addition, the results of two dedicated detection models show improved accuracy compared to the DOTA-trained model.

Estimation of Lower Limb Joint Angles and Joint Moments during Different Locomotive Activities Using the Inertial Measurement Units and a Hybrid Deep Learning Model.

Using inertial measurement units (IMUs) to estimate lower limb joint kinematics and kinetics can provide valuable information for disease diagnosis and rehabilitation assessment. To estimate gait parameters using IMUs, model-based filtering approaches have been proposed, such as the Kalman filter and complementary filter. However, these methods require special calibration and alignment of IMUs. The development of deep learning algorithms has facilitated the application of IMUs in biomechanics as it does not require particular calibration and alignment procedures of IMUs in use. To estimate hip/knee/ankle joint angles and moments in the sagittal plane, a subject-independent temporal convolutional neural network-bidirectional long short-term memory network (TCN-BiLSTM) model was proposed using three IMUs. A public benchmark dataset containing the most representative locomotive activities in daily life was used to train and evaluate the TCN-BiLSTM model. The mean Pearson correlation coefficient of joint angles and moments estimated by the proposed model reached 0.92 and 0.87, respectively. This indicates that the TCN-BiLSTM model can effectively estimate joint angles and moments in multiple scenarios, demonstrating its potential for application in clinical and daily life scenarios.

Fruit 360 classification based on the convolutional neural network

This research paper focuses on the Fruit360 Classification challenge, a task aimed at developing a fruit classification model capable of accurately identifying various fruits and distinguishing them from each other. In this study, the Fruit360 dataset is used, consisting of 90380 images of 131 fruits and vegetable classes. Prior to training the CNN model, the images are preprocessed by resizing, normalizing, and augmenting them. The authors employ a pre-trained CNN model called ResNet-50 using the PyTorch deep learning framework and add a custom fully connected layer on top to adapt the model to the specific classification task. The authors conclude that the proposed model achieved excellent performance on the Fruit360 dataset. The study highlights the importance of the Fruit360 Classification challenge in advancing the field of computer vision, specifically in the development of deep learning algorithms for image classification tasks. The proposed model has the potential to improve the efficiency and accuracy of fruit classification, which can benefit the fruit industry in terms of enhanced productivity and cost-effectiveness.

Research on the Underlying Principles and Deep Learning Algorithms based on Image Style Conversion Techniques

Image style transformation techniques are an important research area in the field of computer vision, aiming to combine the content of one image with the artistic style of another image to create new images with the content of the input image and the style of the artistic style image. The development of this field has benefited from the rapid development of deep learning algorithms, especially the application of techniques such as Convolutional Neural Networks (CNN) and Generative Adversarial Networks (GAN). At the heart of image style transformation is how to capture both the content and artistic style of the input image. This is typically achieved by using different levels of activation of the convolutional neural network as content and style representations. The content representation captures the object and structural information in the image, while the style representation captures the texture and colour information of the image. To capture the style of an image, a Gram matrix is typically used to measure the correlation between features at a particular level. This matrix representation allows us to capture texture information and colour distribution, thus enabling style migration. The goal of the image style transformation model is to minimise both content loss and style loss to produce synthetic images. Content loss typically uses Euclidean distance to measure the difference between content representations, while style loss involves the distance between Gram matrices. GANs have become an important tool in image style transformation. The generator network is responsible for generating synthetic images and the discriminator network evaluates the realism of the generated images. Through adversarial training, the generator and discriminator networks constantly compete to improve the quality of the generated images. Finally, this study highlights the potential applications of image style conversion techniques in the fields of art creation, image editing and virtual reality, and proposes directions for future research on deep learning algorithms and underlying principles to further improve the efficiency and quality of image style conversion techniques. The continuous development of this field will bring new opportunities for image processing and creative applications.

Convolutional Neural Network for Predicting Failure Type in Concrete Cylinders During Compression Testing

Cracks in concrete cause structural damage, and it is important to identify and classify them. The objective of the research was to describe the behavior and predict the type of failure in concrete cylinders using convolutional neural networks. The methodology consisted of creating a database of 2650 images of failure types in concrete cylinders tested in compression at the Laboratory of Testing and Strength of Materials of the National University of Jaen, Cajamarca, Peru. To identify cracks on the concrete surface, the database was divided into training (60%), validation (20%), and testing (20%), and a transfer learning approach was developed using the MobileNet, DenseNet121, ResNet50, and VGG16 algorithms from the Keras library, programmed in Python. To validate the performance of each model, the following indicators were used: recall, precision, and F1 score. The results show that the models studied correctly classified the type of failure in concrete with accuracies of 96, 91, 86, and 90%, with the MobileNet algorithm being the best predictor with 96%. The novelty of the study was the development of deep learning algorithms with different architectures that can be used in structural health assessment as an automated and reliable method compared to traditional ones. In addition, these trained algorithms can be used as source code in drones for structural monitoring. Doi: 10.28991/CEJ-2023-09-09-01 Full Text: PDF

Development of Deep Learning Algorithms for Improved Facial Recognition in Security Applications

This research aims to develop artificial intelligence (AI) algorithms in the context of facial recognition with a focus on increasing accuracy in difficult environmental conditions. Although facial recognition technology has made great progress, challenges such as poor lighting, variations in facial expressions, and head rotation are still problems that must be overcome. The research methodology involved collecting a wide dataset covering a wide variety of faces under various environmental conditions. This data is then processed and its features are extracted using computer image processing techniques. Furthermore, several deep neural network architectures, such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), were developed, trained, and evaluated for face recognition tasks. The expected result is the development of an AI algorithm that is able to overcome challenges in facial recognition with higher accuracy than existing methods. In particular, significant improvements in facial recognition accuracy are expected especially under low lighting conditions and variations in facial expressions. This research has a major impact in a variety of security applications, such as border surveillance, building access control, and corporate security. With higher facial recognition accuracy, security risks can be significantly reduced, resulting in safer and more efficient security solutions. In conclusion, this research aims to bring innovation in facial recognition technology through advanced AI approaches, with the potential to improve security in various contexts.

MIA is an open-source standalone deep learning application for microscopic image analysis

In recent years, the amount of data generated by imaging techniques has grown rapidly, along with increasing computational power and the development of deep learning algorithms. To address the need for powerful automated image analysis tools for a broad range of applications in the biomedical sciences, the Microscopic Image Analyzer (MIA) was developed. MIA combines a graphical user interface that obviates the need for programming skills with state-of-the-art deep-learning algorithms for segmentation, object detection, and classification. It runs as a standalone, platform-independent application and uses open data formats, which are compatible with commonly used open-source software packages. The software provides a unified interface for easy image labeling, model training, and inference. Furthermore, the software was evaluated in a public competition and performed among the top three for all tested datasets.

Children’s dental panoramic radiographs dataset for caries segmentation and dental disease detection

When dentists see pediatric patients with more complex tooth development than adults during tooth replacement, they need to manually determine the patient’s disease with the help of preoperative dental panoramic radiographs. To the best of our knowledge, there is no international public dataset for children’s teeth and only a few datasets for adults’ teeth, which limits the development of deep learning algorithms for segmenting teeth and automatically analyzing diseases. Therefore, we collected dental panoramic radiographs and cases from 106 pediatric patients aged 2 to 13 years old, and with the help of the efficient and intelligent interactive segmentation annotation software EISeg (Efficient Interactive Segmentation) and the image annotation software LabelMe. We propose the world’s first dataset of children’s dental panoramic radiographs for caries segmentation and dental disease detection by segmenting and detecting annotations. In addition, another 93 dental panoramic radiographs of pediatric patients, together with our three internationally published adult dental datasets with a total of 2,692 images, were collected and made into a segmentation dataset suitable for deep learning.

SIVED: A SAR Image Dataset for Vehicle Detection Based on Rotatable Bounding Box

The research and development of deep learning methods are heavily reliant on large datasets, and there is currently a lack of scene-rich datasets for synthetic aperture radar (SAR) image vehicle detection. To address this issue and promote the development of SAR vehicle detection algorithms, we constructed the SAR Image dataset for VEhicle Detection (SIVED) using Ka, Ku, and X bands of data. Rotatable bounding box annotations were employed to improve positioning accuracy, and an algorithm for automatic annotation was proposed to improve efficiency. The dataset exhibits three crucial properties: richness, stability, and challenge. It comprises 1044 chips and 12,013 vehicle instances, most of which are situated in complex backgrounds. To construct a baseline, eight detection algorithms are evaluated on SIVED. The experimental results show that all detectors achieved high mean average precision (mAP) on the test set, highlighting the dataset’s stability. However, there is still room for improvement in the accuracy with respect to the complexity of the background. In summary, SIVED fills the gap in SAR image vehicle detection datasets and demonstrates good adaptability for the development of deep learning algorithms.

A Study on the Development of Deep Learning Algorithm for Determining External Quality of Welded Parts Using Transfer Learning

Recently, deep learning has been applied to various welding techniques, such as laser welding, gas metal arc welding (GMAW), and resistance spot welding, and research on automation and quality prediction is being conducted. Even for GMAW, many researchers have attempted to predict quality through X-ray, current, and voltage measurements. If judgment in real time is not necessary, it is most effective to judge the quality of a welded part using an exterior image. Therefore, in this study, a welded appearance quality judgment model was analyzed using image deep learning. Welding defects were classified into pores, overlaps, craters, melting of the base material, cracks, and undercuts, and were divided into 7 categories including normal ones. In constructing the deep-learning model, transfer learning was performed using existing networks, such as ResNet and AlexNet. To improve the accuracy of deep learning, tests were performed while the optimization technique, maximum number of epochs, and minibatch size were changed. It was confirmed that the accuracy of weld defect prediction improved as the minibatch size increased, and the stochastic gradient descent model had the highest accuracy. Increasing the number of data for learning should make the technique of using images to judge the quality of GMAW welds using the proposed model more widely applicable.

A Deep Learning Approach for Histology-Based Nucleus Segmentation and Tumor Microenvironment Characterization

Microscopic examination of pathology slides is essential to disease diagnosis and biomedical research. However, traditional manual examination of tissue slides is laborious and subjective. Tumor whole-slide image (WSI) scanning is becoming part of routine clinical procedures and produces massive data that capture tumor histologic details at high resolution. Furthermore, the rapid development of deep learning algorithms has significantly increased the efficiency and accuracy of pathology image analysis. In light of this progress, digital pathology is fast becoming a powerful tool to assist pathologists. Studying tumor tissue and its surrounding microenvironment provides critical insight into tumor initiation, progression, metastasis, and potential therapeutic targets. Nucleus segmentation and classification are critical to pathology image analysis, especially in characterizing and quantifying the tumor microenvironment (TME). Computational algorithms have been developed for nucleus segmentation and TME quantification within image patches. However, existing algorithms are computationally intensive and time consuming for WSI analysis. This study presents Histology-based Detection using Yolo (HD-Yolo), a new method that significantly accelerates nucleus segmentation and TME quantification. We demonstrate that HD-Yolo outperforms existing WSI analysis methods in nucleus detection, classification accuracy, and computation time. We validated the advantages of the system on 3 different tissue types: lung cancer, liver cancer, and breast cancer. For breast cancer, nucleus features by HD-Yolo were more prognostically significant than both the estrogen receptor status by immunohistochemistry and the progesterone receptor status by immunohistochemistry. The WSI analysis pipeline and a real-time nucleus segmentation viewer are available at https://github.com/impromptuRong/hd_wsi.

Deforestation detection using a spatio-temporal deep learning approach with synthetic aperture radar and multispectral images

Deforestation is a global change driver that contributes to atmospheric carbon emissions, causes biodiversity loss and ecosystem services degradation. Usually, this process has been quantified and monitored using remote sensing. The development of deep learning algorithms, as well as satellite images with a higher spatial and temporal resolution has improved the capabilities to monitor deforestation. The objective of this study was to evaluate the potential of a spatio-temporal deep learning algorithm, the U-Net 3D, with multispectral and synthetic aperture radar images to detect deforestation in a tropical rainforest in Southeast Mexico between 2019 and 2020. The U-Net 3D inputs were 147 squared areas with 256 pixels by side that contained, at least, one deforested area. These inputs consisted of: (1) four cloudless composites (Feb–Apr 2019, May–Sep 2019, Oct 2019–Jan 2020 and Feb–Apr 2020) with six bands each (four Sentinel-2 bands and two Sentinel-1), and (2) deforested area polygons delineated by visual interpretation. This classification had three classes: no deforestation, old-growth forest loss and secondary forest / plantation loss. The classification map obtained an overall accuracy of 0.97 and an average F1-score of 0.94. The corrected area estimates were 3 195.26 ± 1 132.88 ha (0.48 ± 0.17 %) for the old-growth forest loss and 4 234.85 ± 1 912.01 ha (0.64 ± 0.29 %) for the secondary forest / plantation loss. Most of the classification errors were found in borders between classes, caused by the confusion of secondary forest / plantation loss with herbaceous cover loss or associated with artifacts in the images.