A social image recommendation system based on deep reinforcement learning

Today, due to the expansion of the Internet and social networks, people are faced with a vast amount of dynamic information. To mitigate the issue of information overload, recommender systems have become pivotal by analyzing users' activity histories to discern their interests and preferences. However, most available social image recommender systems utilize a static strategy, meaning they do not adapt to changes in user preferences. To overcome this challenge, our paper introduces a dynamic image recommender system that leverages a deep reinforcement learning (DRL) framework, enriched with a novel set of features including emotion, style, and personality. These features, uncommon in existing systems, are instrumental in crafting a user's characteristic vector, offering a personalized recommendation experience. Additionally, we overcome the challenge of state representation definition in reinforcement learning by introducing a new state representation. The experimental results show that our proposed method, compared to some related works, significantly improves Recall@k and Precision@k by approximately 7%-10% (for the top 100 images recommended) for personalized image recommendation.

A cascade transformer-based model for 3D dose distribution prediction in head and neck cancer radiotherapy

Objective. Radiation therapy is one of the primary methods used to treat cancer in the clinic. Its goal is to deliver a precise dose to the planning target volume while protecting the surrounding organs at risk (OARs). However, the traditional workflow used by dosimetrists to plan the treatment is time-consuming and subjective, requiring iterative adjustments based on their experience. Deep learning methods can be used to predict dose distribution maps to address these limitations.Approach. The study proposes a cascade model for OARs segmentation and dose distribution prediction. An encoder-decoder network has been developed for the segmentation task, in which the encoder consists of transformer blocks, and the decoder uses multi-scale convolutional blocks. Another cascade encoder-decoder network has been proposed for dose distribution prediction using a pyramid architecture. The proposed model has been evaluated using an in-house head and neck cancer dataset of 96 patients and OpenKBP, a public head and neck cancer dataset of 340 patients.Main results. The segmentation subnet achieved 0.79 and 2.71 for Dice and HD95 scores, respectively. This subnet outperformed the existing baselines. The dose distribution prediction subnet outperformed the winner of the OpenKBP2020 competition with 2.77 and 1.79 for dose and dose-volume histogram scores, respectively. Besides, the end-to-end model, including both subnets simultaneously, outperformed the related studies.Significance. The predicted dose maps showed good coincidence with ground-truth, with a superiority after linking with the auxiliary segmentation task. The proposed model outperformed state-of-the-art methods, especially in regions with low prescribed doses. The codes are available athttps://github.com/GhTara/Dose_Prediction.

A generalised deep meta-learning model for automated quality control of cardiovascular magnetic resonance images

BACKGROUND AND OBJECTIVES

Cardiovascular magnetic resonance (CMR) imaging is a powerful modality in functional and anatomical assessment for various cardiovascular diseases. Sufficient image quality is essential to achieve proper diagnosis and treatment. A large number of medical images, the variety of imaging artefacts, and the workload of imaging centres are amongst the factors that reveal the necessity of automatic image quality assessment (IQA). However, automated IQA requires access to bulk annotated datasets for training deep learning (DL) models. Labelling medical images is a tedious, costly and time-consuming process, which creates a fundamental challenge in proposing DL-based methods for medical applications. This study aims to present a new method for CMR IQA when there is limited access to annotated datasets.

METHODS

The proposed generalised deep meta-learning model can evaluate the quality by learning tasks in the prior stage and then fine-tuning the resulting model on a small labelled dataset of the desired tasks. This model was evaluated on the data of over 6,000 subjects from the UK Biobank for five defined tasks, including detecting respiratory motion, cardiac motion, Aliasing and Gibbs ringing artefacts and images without artefacts.

RESULTS

The results of extensive experiments show the superiority of the proposed model. Besides, comparing the model's accuracy with the domain adaptation model indicates a significant difference by using only 64 annotated images related to the desired tasks.

CONCLUSION

The proposed model can identify unknown artefacts in images with acceptable accuracy, which makes it suitable for medical applications and quality assessment of large cohorts. CODE AVAILABILITY: https://github.com/HosseinSimchi/META-IQA-CMRImages.

Prediction of Acute Kidney Injury After Cardiac Surgery Using Interpretable Machine Learning

Background

Acute kidney injury (AKI) is a complication that occurs for various reasons after surgery, especially cardiac surgery. This complication can lead to a prolonged treatment process, increased costs, and sometimes death. Prediction of postoperative AKI can help anesthesiologists to implement preventive and early treatment strategies to reduce the risk of AKI.

Objectives

This study tries to predict postoperative AKI using interpretable machine learning models.

Methods

For this study, the information of 1435 patients was collected from multiple centers. The gathered data are in six categories: demographic characteristics and type of surgery, past medical history (PMH), drug history (DH), laboratory information, anesthesia and surgery information, and postoperative variables. Machine learning methods, including support vector machine (SVM), multilayer perceptron (MLP), decision tree (DT), random forest (RF), logistic regression, XGBoost, and AdaBoost, were used to predict postoperative AKI. Local interpretable model-agnostic explanations (LIME) and the Shapley methods were then leveraged to check the interpretability of models.

Results

Comparing the area under the curves (AUCs) obtained for different machine learning models show that the RF and XGBoost methods with values of 0.81 and 0.80 best predict postoperative AKI. The interpretations obtained for the machine learning models show that creatinine (Cr), cardiopulmonary bypass time (CPB time), blood sugar (BS), and albumin (Alb) have the most significant impact on predictions.

Conclusions

The treatment team can be informed about the possibility of postoperative AKI before cardiac surgery using machine learning models such as RF and XGBoost and adjust the treatment procedure accordingly. Interpretability of predictions for each patient ensures the validity of obtained predictions.

Medical imaging and computational image analysis in COVID-19 diagnosis: A review

Coronavirus disease (COVID-19) is an infectious disease caused by a newly discovered coronavirus. The disease presents with symptoms such as shortness of breath, fever, dry cough, and chronic fatigue, amongst others. The disease may be asymptomatic in some patients in the early stages, which can lead to increased transmission of the disease to others. This study attempts to review papers on the role of imaging and medical image computing in COVID-19 diagnosis. For this purpose, PubMed, Scopus and Google Scholar were searched to find related studies until the middle of 2021. The contribution of this study is four-fold: 1) to use as a tutorial of the field for both clinicians and technologists, 2) to comprehensively review the characteristics of COVID-19 as presented in medical images, 3) to examine automated artificial intelligence-based approaches for COVID-19 diagnosis, 4) to express the research limitations in this field and the methods used to overcome them. Using machine learning-based methods can diagnose the disease with high accuracy from medical images and reduce time, cost and error of diagnostic procedure. It is recommended to collect bulk imaging data from patients in the shortest possible time to improve the performance of COVID-19 automated diagnostic methods.

A View Transformation Model Based on Sparse and Redundant Representation for Human Gait Recognition

Background

Human gait as an effective behavioral biometric identifier has received much attention in recent years. However, there are challenges which reduce its performance. In this work we aim at improving performance of gait systems under variations in view angles, which present one of the major challenges to gait algorithms.

Methods

We propose employment of a view transformation model based on sparse and redundant (SR) representation. More specifically, our proposed method trains a set of corresponding dictionaries for each viewing angle, which are then used in identification of a probe. In particular, the view transformation is performed by first obtaining the SR representation of the input image using the appropriate dictionary, then multiplying this representation by the dictionary of destination angle to obtain a corresponding image in the intended angle.

Results

Experiments performed using CASIA Gait Database, Dataset B, support the satisfactory performance of our method. It is observed that in most tests, the proposed method outperforms the other methods in comparison. This is especially the case for large changes in the view angle, as well as the average recognition rate.

Conclusion

A comparison with state-of-the-art methods in the literature showcases the superior performance of the proposed method, especially in the case of large variations in view angle.

An unsupervised hierarchical approach for automatic intra-retinal cyst segmentation in spectral-domain optical coherence tomography images

PURPOSE

Intra-retinal cyst (IRC) is a symptom of macular disorders that occurs due to retinal blood vessel damage and fluid leakage to the macula area. These abnormalities are efficiently visualized using optical coherence tomography (OCT) imaging. These patients need to be regularly monitored for the presence and changes of IRC regions. Thus, automatic segmentation of IRCs can be beneficial to investigate disease progression.

METHODS

In this study, automatic IRC segmentation is accomplished by building three different masks in three unsupervised segmentation levels of a hierarchical framework. In the first level, the ROI-mask (R-mask) is built, and the retina area is cropped based on this mask. In the second level, the prune-mask (P-mask) is built, and the searching space is significantly reduced toward the target objects using this mask; and finally in the third level, by applying the Markov random field (MRF) model and employing intensity and contextual information, the cyst mask (C-mask) is extracted.

RESULTS

The proposed method is evaluated on three datasets including OPTIMA, UMN, and KERMANY datasets. The experimental results showed that the proposed method is effective with a mean dice coefficient rate of 0.74, 0.75 and 0.79 by the intersection of ground truths on the OPTIMA, UMN and KERMANY datasets, respectively.

CONCLUSION

The proposed method outperforms the state-of-the-art methods on the OPTIMA and UMN datasets while achieving comparable results to the most recently proposed method on the KERMANY dataset.

Respiratory Motion Prediction Using Deep Convolutional Long Short-Term Memory Network

Background

Pulmonary movements during radiation therapy can cause damage to healthy tissues. It is necessary to adapt treatment planning based on tumor motion to avoid damage to healthy tissues. A range of approaches has been proposed to monitor the issue. A treatment planning based on fourdimensional computed tomography (4D CT) images can be addressed as one of the most achievable options. Although several methods proposed to predict pulmonary movements based on mathematical algorithms, the use of deep artificial neural networks has recently been considered.

Methods

In the current study, convolutional long shortterm memory networks are applied to predict and generate images throughout the breathing cycle. A total of 3295 CT images of six patients in three different views was considered as reference images. The proposed method was evaluated in six experiments based on a leaveonepatientout method similar to crossvalidation.

Results

The weighted average results of the experiments in terms of the rootmeansquared error and structural similarity index measure are 9 × 10^-3 and 0.943, respectively.

Conclusion

Utilizing the proposed method, because of its generative nature, which results in the generation of CT images during the breathing cycle, improves the radiotherapy treatment planning in the lack of access to 4D CT images.

A Multimodal Fusion Approach for Bullet Identification Systems

In the field of forensic science, bullet identification is based on the fact that firing the cartridge from a barrel leaves exclusive microscopic striation on the fired bullets as the fingerprint of the firearm. The bullet identification methods are categorized in 2-D and 3-D based on their image acquisition techniques. In this study, we focus on 2-D optical images using a multimodal technique and propose several distinct methods as its modalities. The proposed method uses a multimodal rule-based linear weighted fusion approach which combines the semantic level decisions from different modalities with a linear technique that its optimized modalities weights have been identified by the genetic algorithm. The proposed approach was applied on a dataset, which includes 180 2-D bullet images fired from 90 different AK-47 barrels. The experimentations showed that our approach attained better results compared to common methods in the field of bullet identification.

Automatic segmentation of abnormal capillary nonperfusion regions in optical coherence tomography angiography images using marker-controlled watershed algorithm

Diabetic retinopathy (DR) is one of the most complications of diabetes. It is a progressive disease leading to significant vision loss in the patients. Abnormal capillary nonperfusion (CNP) regions are one of the important characteristics of DR increasing with its progression. Therefore, automatic segmentation and quantification of abnormal CNP regions can be helpful to monitor the patient's treatment process. We propose an automatic method for segmentation of abnormal CNP regions on the superficial and deep capillary plexuses of optical coherence tomography angiography (OCTA) images using the marker-controlled watershed algorithm. The proposed method has three main steps. In the first step, original images are enhanced using the vesselness filter and then foreground and background marker images are computed. In the second step, abnormal CNP region candidates are segmented using the marker-controlled watershed algorithm, and in the third step, the candidates are modeled using an undirected weighted graph and finally, by applying merging and removing procedures correct abnormal CNP regions are identified. The proposed method was evaluated on a dataset with 36 normal and diabetic subjects using the ground truth obtained by two observers. The results show the proposed method outperformed some of the state-of-the-art methods on the superficial and deep capillary plexuses according to the most important metrics.

Blind Detection of Region Duplication Forgery Using Fractal Coding and Feature Matching

Digital image forgery detection is important because of its wide use in applications such as medical diagnosis, legal investigations, and entertainment. Copy-move forgery is one of the famous techniques, which is used in region duplication. Many of the existing copy-move detection algorithms cannot effectively blind detect duplicated regions that are made by powerful image manipulation software like Photoshop. In this study, a new method is proposed for blind detecting manipulations in digital images based on modified fractal coding and feature vector matching. The proposed method not only detects typical copy-move forgery, but also finds multiple copied forgery regions for images that are subjected to rotation, scaling, reflection, and a mixture of these postprocessing operations. The proposed method is robust against tampered images undergoing attacks such as Gaussian blurring, contrast scaling, and brightness adjustment. The experimental results demonstrated the validity and efficiency of the method.

A New Approach for Copy-Move Detection Based on Improved Weber Local Descriptor

One of the most common image tampering techniques is copy-move; in this technique, one or more parts of the image are copied and pasted in another area of the image. Recently, various methods have been proposed for copy-move detection; however, many of these techniques are not robust to additional changes like geometric transformation, and they are failed to be useful for detecting small copied areas. In this paper, a new method based on point descriptors which are derived from the integration of textural feature-based Weber law and statistical features of the image is presented. In this proposed approach, modified multiscale version of Weber local descriptor is presented to make the method robust versus geometric transformation and detect small copied areas. The results of the experiments showed that our method can detect small copied areas and copy-move tampered images which are influenced by rotation, scaling, noise addition, compression, blurring, and mirroring.

SOMM: A new service oriented middleware for generic wireless multimedia sensor networks based on code mobility

Although much research in the area of Wireless Multimedia Sensor Networks (WMSNs) has been done in recent years, the programming of sensor nodes is still time-consuming and tedious. It requires expertise in low-level programming, mainly because of the use of resource constrained hardware and also the low level API provided by current operating systems. The code of the resulting systems has typically no clear separation between application and system logic. This minimizes the possibility of reusing code and often leads to the necessity of major changes when the underlying platform is changed. In this paper, we present a service oriented middleware named SOMM to support application development for WMSNs. The main goal of SOMM is to enable the development of modifiable and scalable WMSN applications. A network which uses the SOMM is capable of providing multiple services to multiple clients at the same time with the specified Quality of Service (QoS). SOMM uses a virtual machine with the ability to support mobile agents. Services in SOMM are provided by mobile agents and SOMM also provides a t space on each node which agents can use to communicate with each other.