Automatic Classification Algorithm Research Articles

Enhancing Brain Tumor Classification: A CNN-Based Approach with InceptionV3 and Xception

Brain tumors are among the most aggressive and deadly diseases, with a very short life expectancy at the highest grade. To combat this, early detection and treatment are crucial.In this approach, MRI images are used to analyze brain abnormalities. The manual investigation of brain tumor classification is a time-consuming task, and there might be possibilities of human errors. Hence, accurate analysis in a short span of time is essential. This approach presents the automatic brain tumor classification algorithm using a highly accurate Convolutional Neural Network (CNN), InceptionV3 and Xception algorithm for classification of Glioma, Meningioma and No tumor. The brain part is initially segmented by a thresholding approach followed by a morphological operation. The brain MRI is classified using CNN, Inceptionv3, and Xceptionv3 algorithms. The system's performance is evaluated using precision, recall, F1 score, and accuracy parameters.

Intelligent in-process enhancement technique for machining efficiency in CNC machine tools based on spindle power

In the realm of mechanical machining, adaptive machining techniques offer an efficient method. However, existing adaptive machining technologies cannot automatically identify the type of workpiece to be machined, nor can they set different control targets based on tool load-bearing capacity. This often requires manual and cumbersome operations, making the application of adaptive machining technology difficult to popularize. To address these challenges, this paper proposes an intelligent and efficient adaptive machining method. Specifically, a workpiece automatic classification algorithm based on machine tool spindle power is introduced. This algorithm classifies the machined workpieces according to the collected spindle power data, enhancing the convenience of adaptive machining technology. Also based on spindle power data, a time-series segmentation method is proposed, dividing spindle power into different subsequences according to the tools used in machining, which enhances the accuracy of adaptive machining technology. Furthermore, an adaptive feed rate control algorithm is designed based on fuzzy control theory, realizing intelligent and efficient adaptive machining of workpieces to improve machining efficiency. Finally, experiments are conducted to validate the effectiveness of the intelligent and efficient adaptive machining method. The proposed method effectively addresses the drawbacks of existing adaptive machining method, which require complex parameter settings. It simplifies the operational process, enhances the intelligence, and improves the practical applicability of adaptive machining techniques. This represents an innovative contribution to the field of adaptive machining techniques.

The Cloud Indicator: A novel algorithm for automatic detection and classification of clouds using airborne in situ observations

Airborne in situ observations of aerosols and clouds provide relevant data sets crucial to improving the understanding of cloud microphysical processes and reducing uncertainties connected to future climate predictions. However, the selection and classification of cloud sequences in such data sets can be very time-consuming if done manually, and criteria used by the community are numerous and prone to misclassification, especially when coarse aerosol particles (> 1 μm diameter) are present.In this study, we present the Cloud Indicator, a novel algorithm that automatically detects and classifies measurement periods inside clouds. The Cloud Indicator was developed using data from three international airborne field campaigns, including ATom (Atmospheric Tomography; 2016–2018), A-LIFE (Absorbing aerosol layers in a changing climate: aging, lifetime and dynamics; 2017), and FIREX-AQ (Fire Influence on Regional to Global Environments Experiment and Air Quality; 2019). The algorithm utilizes size distribution measurements, combined with measurements of relative humidity and temperature, to automatically detect flight sequences in clouds and classify the cloud type. As an additional criterion for the Cloud Indicator, we established the cloud-aerosol volume factor fCA to ensure a precise and robust distinction between clouds and aerosol layers such as mineral dust or biomass burning thereby reducing misclassifications. The Cloud Indicator algorithm was developed with data from a second-generation Cloud, Aerosol, and Precipitation Spectrometer (CAPS) which was calibrated with a novel calibration procedure – introduced in this study - that results in a refractive index independent calibration and works also in the field. However, the Cloud Indicator algorithm is not restricted specifically to the CAPS and thus allows its application to a variety of in situ instruments that measure coarse aerosol and cloud particle size.Case studies from ATom and A-LIFE demonstrate the ability of the Cloud Indicator to precisely screen (airborne) in situ data sets for clouds. The algorithm separates the data set into cloud-free periods, Aerosol-Cloud Transition Regime (ACTR), liquid clouds, clouds in the Mixed-Phase Temperature Regime (MPTR), and cirrus clouds. The unique ability of the Cloud Indicator to successfully differentiate between layers of enhanced coarse-mode aerosol concentrations and clouds is demonstrated by measurements in a complex mixture of clouds embedded into a mineral dust layer during the A-LIFE. The empirically derived parameter thresholds of the Cloud Indicator are in good agreement with values in the literature.

Brain Tumor Detection Using Deep Learning

The brain tumor is one of the most dangerous, common and aggressive diseases which leads to a very short life expectancy at the highest grade. Thus, to prevent life from such disease, early recognition, and fast treatment is an essential step. In this approach, MRI images are used to analyze brain abnormalities. The manual investigation of brain tumor classification is a time-consuming task and there might have possibilities of human errors. Hence accurate analysis in a tiny span of time is an essential requirement. In this approach, the automatic brain tumor classification algorithm using a highly accurate Convolutional Neural Network (CNN) algorithm is presented. Initially, the brain part is segmented by thresholding approach followed by a morphological operation. The brain MRI is classified using CNN, Inceptionv3, and Xceptionv3 algorithms. The performance of the system is evaluated using precision, recall, F1 score and accuracy parameter

Automatic and real-time tissue sensing for autonomous intestinal anastomosis using hybrid MLP-DC-CNN classifier-based optical coherence tomography.

Anastomosis is a common and critical part of reconstructive procedures within gastrointestinal, urologic, and gynecologic surgery. The use of autonomous surgical robots such as the smart tissue autonomous robot (STAR) system demonstrates an improved efficiency and consistency of the laparoscopic small bowel anastomosis over the current da Vinci surgical system. However, the STAR workflow requires auxiliary manual monitoring during the suturing procedure to avoid missed or wrong stitches. To eliminate this monitoring task from the operators, we integrated an optical coherence tomography (OCT) fiber sensor with the suture tool and developed an automatic tissue classification algorithm for detecting missed or wrong stitches in real time. The classification results were updated and sent to the control loop of STAR robot in real time. The suture tool was guided to approach the object by a dual-camera system. If the tissue inside the tool jaw was inconsistent with the desired suture pattern, a warning message would be generated. The proposed hybrid multilayer perceptron dual-channel convolutional neural network (MLP-DC-CNN) classification platform can automatically classify eight different abdominal tissue types that require different suture strategies for anastomosis. In MLP, numerous handcrafted features (∼1955) were utilized including optical properties and morphological features of one-dimensional (1D) OCT A-line signals. In DC-CNN, intensity-based features and depth-resolved tissues' attenuation coefficients were fully exploited. A decision fusion technique was applied to leverage the information collected from both classifiers to further increase the accuracy. The algorithm was evaluated on 69,773 testing A-line data. The results showed that our model can classify the 1D OCT signals of small bowels in real time with an accuracy of 90.06%, a precision of 88.34%, and a sensitivity of 87.29%, respectively. The refresh rate of the displayed A-line signals was set as 300 Hz, the maximum sensing depth of the fiber was 3.6 mm, and the running time of the image processing algorithm was ∼1.56 s for 1,024 A-lines. The proposed fully automated tissue sensing model outperformed the single classifier of CNN, MLP, or SVM with optimized architectures, showing the complementarity of different feature sets and network architectures in classifying intestinal OCT A-line signals. It can potentially reduce the manual involvement of robotic laparoscopic surgery, which is a crucial step towards a fully autonomous STAR system.

A novel automatic detection and classification algorithm for elderly cognitive impairment using CSVM

Alzheimer’s disease (AD) is a degenerative disease with insidious onset and chronic nervous system, which can be roughly classified into three stages: preclinical AD, mild cognitive impairment (MCI), and clinical AD. Given that timely treatment of MCI can delay the course of AD, identifying what stage of the disease is at the time of early diagnosis is critical in clinical practice. Based on the survey data of elderly residents in Changsha community, this paper constructs the cognitive impairment data set and explores the application of the emerging classification algorithm Convolutional Support Vector Machine(CSVM) in identifying elderly cognitive impairment, a small data classification problem. CSVM is an improved Support Vector Machine classification(SVC) algorithm, which uses the convolution filtering idea of the Convolutional Neural Networks(CNN) algorithm to preprocess the nonlinearly separable sample set to improve the classification ability of SVM. In addition, the optimal convolution filter (solution) is found by the Simplified Swarm Optimization(SSO) algorithm and orthogonal array experimental design. A variable update amplitude method is also proposed to optimize the SSO update strategy further. Aiming at the problem of an unbalanced sample set, a stratified sampling method is used to divide and cross-validate the sample data, and a variety of evaluation indicators are used to evaluate the performance of the model. Experimental results show that the data set processed by a specific convolution filter can significantly improve the classification performance of linear SVM. The accurate differentiation of normal cognition(NC), MCI, and AD helps intervene in adjuvant treatment for potential patients timely.

The highlights of kidney transplantation in2023

In 2023, significant advances were made in various areas of kidney transplantation. Firstly, the use of a balanced crystalloid solution in the recipient appears to prevent the delay in graft function, unlike hypothermia in the donor and normothermic pulsatile perfusion. Understanding the pathophysiology of humoral rejection has progressed, highlighting the major role of HLA class II molecules and innate immune cells (NK and monocytes expressing FCGR3A). An automatic Banff classification algorithm has been developed to better categorize biopsies in currently known diagnoses. CXCL10, combined with other variables, seems effective in ruling out rejection, but its role in routine care is yet to be defined. Regarding cytomegalovirus (CMV), letermovir has been proven effective in preventing CMV disease in D+R- patients, with fewer hematological side effects. For R+ patients, monitoring CMV-specific T-cell immunity is suggested to reduce the duration of antiviral prophylaxis. The only innovation in immunosuppression is imlifidase for highly sensitized patients, guided by French recommendations. A new equation for glomerular filtration rate measurement has been developed for kidney transplant recipients, performing well across various analyzed stratifications. Finally, xenotransplantation is making a comeback this year, generating hope. However, the description of early humoral rejections involving innate immune cells indicates that adjustments are still needed before considering its widespread deployment.

Experimental characterization on fracture behavior of UHPMC under small-scale sample tensile testing: Acoustic emission monitoring and digital image correlation

Manufactured sand (MS) was used to partially replace the traditional quartz sand (QS) to prepare a novel ultra-high-performance manufactured sand concrete (UHPMC). To better understand the fracture damage mechanism of UHPMC, small-scale sample tensile tests were conducted to investigate the tensile behavior of three types of UHPMC (low strain hardening, low strain softening and high strain softening). Acoustic emission (AE) and digital image correlation (DIC) are employed simultaneously to monitor and assess the damage characteristics of UHPMC. The results indicate that the classical AE characteristic parameters (ringing counts, duration and b-value) provide a satisfactory result in identifying the feature stages of low strain hardening and high strain softening. Based on the Gaussian mixture model (GMM) and support vector machine (SVM), an automatic probability classification algorithm has been developed that achieves the quantitative categorization of three UHPMC cracking types (shear or tensile cracks). Among them, tensile damage is dominant during the entire loading process, whereas shear damage, including aggregate dislocation and fiber debonding and pullout, accumulates during the hardening and softening stages. DIC visualization and AE source localization complement each other in identifying crack distribution, contributing to further revealing the tensile damage mechanism of UHPMC. The results obtained in this study can provide data and theoretical support for fracture damage characterization and health monitoring of ultra-high-performance concrete materials under tensile conditions.

An IoT and Machine Learning-based Neonatal Sleep Stage Classification

Sleep, in neonates, is used to access the quality of brain and physical development. Typically, neonatal sleep has been divided into three stages: active sleep (AS), quiet sleep (QS), and intermediate sleep (IS). Polysomnography (PSG) is considered a gold standard to classify sleep. To address this issue, over the past two decades, researchers proposed multiple algorithms for automatic sleep stage classification. These algorithms work achieved outstanding results i.e. quiet sleep detection still, lacks in many aspects. One major drawback of the existing research is amalgamation of awake and active sleep into low voltage irregular (LVI) state. This amalgamation corrupts 40% of the overall EEG signal. For this reason, we proposed an algorithm for neonatal sleep-wake classification using machine learning. The proposed research is divided into three steps. Firstly, the EEG signal was pre-processed using finite impulse response filter to remove the noise and artifacts. Clean EEG signal is then divided into 4560 30-sec segments. Then, twenty prominent EEG features were extracted from time, frequency, and spatial domain. After feature extraction, support vector machine was used for sleep stage classification. The propounded study outperformed all the existing algorithms for sleep-wake classification with a mean accuracy of 83.7%. Four-fold cross-validation was used to validate the overall dataset. Multiple other performance matrices i.e. sensitivity, specificity, Kappa were calculated to prove the efficacy of the proposed study. Statistical results show that the proposed study can be used as a real-time neonatal sleep and Awake classification algorithm, as this did not use prior post-processing techniques.

Strategies for Integrating Red Cultural Resources into Integrated Teaching of Civics and Political Science Courses in Universities and Primary Schools in the Context of Information Age

Abstract The in-depth integration of big data technology and red cultural resources promotes the integration of red cultural resources into the integrated construction of the Civic and Political Science curriculum. In this paper, we use information gained to extract the text features of red cultural resources, combine TF*IDF to calculate the weights of feature items, represent the text to the vector space model, put forward the automatic classification algorithm of decision tree red cultural resources text, and build the framework system of red cultural resources text classification. Teaching stage and political appearance are used as independent variables, respectively, to explore the differences in the teaching effect of Civics courses at different stages of universities, secondary schools, and elementary schools as well as among CPC members, members of the League, and the people, and focuses on analyzing the utilization rate of the red cultural resources of Civics course teaching as well as the students’ participation degree. The results show that only 0.0537 students think that the school functionaries are not working correctly, and there exist 0.4308 students who believe that the classroom teaching is not innovative enough. Combining red cultural resources and the integrated teaching of Civic and Political Science courses requires more excavation of red cultural resources and connotations and the formation of characteristic teaching.

Enhancing Medical Image Classification: A Deep Learning Perspective with Multi Wavelet Transform

Classification of medical images is a very important area of research for both the medical industry and academia. In recent years, automated classification algorithms have become very important in most medical applications, saving time and effort, such as disease detection and diagnostic radiology. Deep learning offers a plethora of advantages when applied to medical image classification, revolutionizing medical diagnosis and patient care. In this study, deep convolutional neural networks (DCNNs) is used to classify medical im-ages and multi-wavelet transform will be applied to extract features. The proposed method aims to improve medical image classification accuracy, thereby assisting healthcare professionals in making more accurate and efficient diagnoses. DCNNs based on the VGG16 model were trained and used in this study. Combining VGG16, a powerful convolutional neural network (CNN), with multiwavelet transform offers several advantages for image processing and analysis tasks, particularly in areas like image classification and feature extraction. To evaluate the performance of the proposed method six publicly available brain tumour MRI datasets are analysed with DCNNs. A fully connected layer is used to categorize the extracted features. According to the results, the deep CNN model combined with the multi-wavelet trans-form achieves an impressive accuracy of 96.43 %. It is evident from this high level of accuracy that the proposed approach is effective in accurately classifying medical images.

Автоматизация классификации сейсмических событий при сейсмомониторинге угольной шахты с использованием машинного обучения

The paper is dedicated to the development of an algorithm for automatic classification of geodynamic processes in the context of monitoring seismic activity in mines using machine learning methods. The importance of classifying geodynamic processes is noted in terms of understanding the nature of seismic phenomena, identifying their sources, assessing the potential hazard as well as their impact on the environment and the infrastructure of underground structures. The paper describes an algorithm for analyzing seismic activity based on the data obtained as the result of recording seismic events using the hardware complex and the GITS2 seismic monitoring software in a coal mine. The paper briefly examines the key artificial intelligence methods used to control and predict hazardous geodynamic phenomena. Particular attention is paid to the development of a machine learning model based on decision trees that demonstrates high accuracy in classifying seismic events. The classification accuracy of the developed model is 98,39% on the training set and 98,41% on the test set. This result indicates the high generalization ability of the model on new data and the absence of overfitting. Testing the algorithm on new data entering the system also confirms the high accuracy in classification of seismic event types with the level of 83–93%. This highlights the efficiency of machine learning methods in mine seismic control. After its trial operation, the developed machine learning model will be implemented in the GITS2 monitoring system, that will allow classification of the incoming seismic events in automatic mode.

Automated algorithm selection using meta-learning and pre-trained deep convolution neural networks

Automated machine learning (AutoML), which explores the automation of various machine learning tasks, has seen significant progress in recent years. One of the fundamental tasks in this field is the selection of effective algorithms for a given dataset. This task is particularly challenging for tabular datasets where, unlike images or text, one cannot use pre-trained embeddings from other datasets. In this study, we present AutoIRAD, a novel meta-learning and vision-based approach for automatic classification algorithm selection for tabular datasets. Our approach is the first to generate image-based representations of entire tabular datasets, enabling us to model the features and interactions of all the dataset samples. Our image-based representation has two significant advantages. First, training our model across multiple datasets enables it to simultaneously learn from diverse domains and create more generalized and effective dataset representations. This representation can also be effectively applied to previously unseen datasets. Secondly, using images enables us to leverage large pre-trained networks (e.g., VGG). By tuning existing networks rather than training new ones, we can both speed up the training of our model and reduce the needed amounts of training data. An extensive evaluation of 150 datasets shows that AutoIRAD achieves results comparable to the leading state-of-the-art AutoML solution but at a fraction of their computation time.

Scale dependence of forest fragmentation and its climate sensitivity in a semi-arid mountain: Comparing Landsat, Sentinel and Google Earth data

Landscape fragmentation is generally viewed as an indicator of environmental stresses or risks, but the fragmentation intensity assessment also depends on the scale of data and the definition of spatial unit. This study aimed to explore the scale-dependence of forest fragmentation intensity along a moisture gradient in Yinshan Mountain of North China, and to estimate environmental sensitivity of forest fragmentation in this semi-arid landscape. We developed an automatic classification algorithm using simple linear iterative clustering (SLIC) and Gaussian mixture model (GMM), and extracted tree canopy patches from Google Earth images (GEI), with an accuracy of 89.2% in the study area. Then we convert the tree canopy patches to forest category according to definition of forest that tree density greater than 10%, and compared it with forest categories from global land use datasets, FROM-GLC10 and GlobeLand30, with spatial resolutions of 10 m and 30 m, respectively. We found that the FROM-GLC10 and GlobeLand30 datasets underestimated the forest area in Yinshan Mountain by 16.88% and 21.06%, respectively; and the ratio of open forest (OF, 10% < tree coverage < 40%) to closed forest (CF, tree coverage > 40%) areas in the underestimated part was 2:1. The underestimations concentrated in warmer and drier areas occupied mostly by large coverage of OFs with severely fragmented canopies. Fragmentation intensity of canopies positively correlated with spring temperature while negatively correlated with summer precipitation and terrain slope. When summer precipitation was less than 300 mm or spring temperature higher than 4 °C, canopy fragmentation intensity rose drastically, while the forest area percentage kept stable. Our study suggested that the spatial configuration, e.g., sparseness, is more sensitive to drought stress than area percentage. This highlights the importance of data resolution and proper fragmentation measurements for forest patterns and environmental interpretation, which is the base of reliable ecosystem predictions with regard to the future climate scenarios.

Data-driven automated classification algorithms for acute health conditions: applying PheNorm to COVID-19 disease.

Automated phenotyping algorithms can reduce development time and operator dependence compared to manually developed algorithms. One such approach, PheNorm, has performed well for identifying chronic health conditions, but its performance for acute conditions is largely unknown. Herein, we implement and evaluate PheNorm applied to symptomatic COVID-19 disease to investigate its potential feasibility for rapid phenotyping of acute health conditions. PheNorm is a general-purpose automated approach to creating computable phenotype algorithms based on natural language processing, machine learning, and (low cost) silver-standard training labels. We applied PheNorm to cohorts of potential COVID-19 patients from 2 institutions and used gold-standard manual chart review data to investigate the impact on performance of alternative feature engineering options and implementing externally trained models without local retraining. Models at each institution achieved AUC, sensitivity, and positive predictive value of 0.853, 0.879, 0.851 and 0.804, 0.976, and 0.885, respectively, at quantiles of model-predicted risk that maximize F1. We report performance metrics for all combinations of silver labels, feature engineering options, and models trained internally versus externally. Phenotyping algorithms developed using PheNorm performed well at both institutions. Performance varied with different silver-standard labels and feature engineering options. Models developed locally at one site also worked well when implemented externally at the other site. PheNorm models successfully identified an acute health condition, symptomatic COVID-19. The simplicity of the PheNorm approach allows it to be applied at multiple study sites with substantially reduced overhead compared to traditional approaches.

Classification of patients with dementia in a late life depression cohort using structural MRI

AbstractBackgroundLate life depression (LLD) is a risk factor for cognitive impairment, with up to 40% of depressed older adults showing persistent cognitive deficits even after symptom remission. Over 20% of Alzheimer’s disease patients exhibit depression‐like symptoms which can represent an early marker of disease (Tsuno et Homma., 2009). We hypothesized that automated classification algorithms trained on structural MRI could predict adjudicated diagnosis of cognitive impairment (CI) in this cohort.MethodOur study evaluated 295 patients with LLD. An adjudication committee consisting of two board‐certified psychiatrists and a clinical neuropsychologist determined ground‐truth labels for patient diagnostic results of CI (N = 145) or No CI (N = 150). Patients underwent T1‐weighted scans, which were segmented with Freesurfer 6.0.1 (freesurfer.net) packaged with fMRIPrep 1.5.8 (fmriprep.org) using the Desikan‐Killiany atlas. Cortical thickness and volumetric data were obtained along with demographic parameters. Freesurfer data was then regressed against estimated total intracranial volume. 11 classifications algorithms implemented in R were trained and tested on 209 and 88 patients, respectively, to classify if a patient has CI. All continuous‐valued data were normalized with respect to the training set mean and standard deviation.ResultTable 1 shows results from multiple classifications algorithms, including high sensitivity, specificity, and accuracy from adaptive boosting algorithms. The latter algorithms iteratively optimize a set of binary classification tree algorithms (Freund et Schapire., 1996).ConclusionVolumetric structural MRI biomarkers trained with machine learning classifiers can adequately categorize depressed older adults with CI which could assist the adjudication process. Future work incorporating structural MRI biomarkers with gold standard clinical data may improve earlier detection of cognitive impairment stemming from a neurodegenerative etiology.

Survey of freshwater mussels using high‐resolution acoustic imaging sonar and deep learning‐based object detection in Lake Izunuma, Japan

Abstract The use of traditional in situ methods for underwater surveys to map freshwater mussel habitats is limited by challenges such as water transparency, depth and high labour demands. In this study, adaptive resolution imaging sonar (ARIS) was applied to monitor mussel distribution and abundance. In contrast to conventional quadrat surveys, this acoustic survey is non‐invasive and enables direct observation of mussels to allow their survival status to be determined in turbid water. ARIS produces high‐quality acoustic images that facilitate the creation of a distribution map for broader‐scale monitoring, especially if paired with deep learning methods such as the YOLOv4 algorithm for automatic mussel detection and classification. ARIS was successfully applied to surveying over 2000 m2 of Lake Izunuma in Miyagi, Japan. In one site, a high mussel abundance of ~0.6 individuals/m2 was detected, while other sites had low densities. The detection model achieved a mean average precision of 0.97. The survey results were used to generate a distribution map of living mussels. This study illustrates the feasibility of using an acoustic video camera with an open‐source deep learning algorithm to monitor mussels and other benthos in turbid water.

From Pulses to Sleep Stages: Towards Optimized Sleep Classification Using Heart-Rate Variability.

More and more people quantify their sleep using wearables and are becoming obsessed in their pursuit of optimal sleep ("orthosomnia"). However, it is criticized that many of these wearables are giving inaccurate feedback and can even lead to negative daytime consequences. Acknowledging these facts, we here optimize our previously suggested sleep classification procedure in a new sample of 136 self-reported poor sleepers to minimize erroneous classification during ambulatory sleep sensing. Firstly, we introduce an advanced interbeat-interval (IBI) quality control using a random forest method to account for wearable recordings in naturalistic and more noisy settings. We further aim to improve sleep classification by opting for a loss function model instead of the overall epoch-by-epoch accuracy to avoid model biases towards the majority class (i.e., "light sleep"). Using these implementations, we compare the classification performance between the optimized (loss function model) and the accuracy model. We use signals derived from PSG, one-channel ECG, and two consumer wearables: the ECG breast belt Polar® H10 (H10) and the Polar® Verity Sense (VS), an optical Photoplethysmography (PPG) heart-rate sensor. The results reveal a high overall accuracy for the loss function in ECG (86.3 %, κ = 0.79), as well as the H10 (84.4%, κ = 0.76), and VS (84.2%, κ = 0.75) sensors, with improvements in deep sleep and wake. In addition, the new optimized model displays moderate to high correlations and agreement with PSG on primary sleep parameters, while measures of reliability, expressed in intra-class correlations, suggest excellent reliability for most sleep parameters. Finally, it is demonstrated that the new model is still classifying sleep accurately in 4-classes in users taking heart-affecting and/or psychoactive medication, which can be considered a prerequisite in older individuals with or without common disorders. Further improving and validating automatic sleep stage classification algorithms based on signals from affordable wearables may resolve existing scepticism and open the door for such approaches in clinical practice.

Automatic classification of heart failure based on Cine-CMR images.

Heart failure (HF) is a serious and complex syndrome with a high mortality rate. In clinical diagnosis, the correct classification of HF is helpful. In our previous work, we proposed a self-supervised learning framework of HF classification (SSLHF) on cine cardiac magnetic resonance images (Cine-CMR). However, this method lacks the integration of three dimensions of spatial information and temporal information. Thus, this study aims at proposing an automatic 4D HF classification algorithm. To construct a 4D classification model, we proposed an extensional framework called 4D-SSLHF. It mainly consists of self-supervised image restoration and HF classification. The image restoration proxy task utilizes three image transformation methods to enhance the exploration of spatial and temporal information in the Cine-CMR. In the classification task, we proposed a Siamese Conv-LSTM network by combining the Siamese network and bi-directional Conv-LSTM to integrate the features of the four dimensions simultaneously. Experimental results on 184 patients from Shanghai Chest Hospital achieved an AUC of 0.8794 and an ACC of 0.8402 in the five-fold cross-validation. Compared with our previous work, the improvements in AUC and ACC were 2.89 % and 1.94 %, respectively. In this study, we proposed a novel self-supervised learning framework named 4D-SSLHF for HF classification based on Cine-CMR. The proposed 4D-SSLHF can mine 3D spatial information and temporal information in Cine-CMR images well and accurately classify different categories of HF. The good classification results show our method's potential to assist physicians in choosing personalized treatment.

Comparison of machine learning models for MRI image-based brain tumor classification and segmentation

Brain tumors have a high-risk factor and are extremely harmful to the human body. With the development of science and technology in recent years, automatic segmentation has become popular in medical diagnosis because it provides higher accuracy than traditional hand segmentation. At present, more and more people start to study and improve it. Due to the non-invasive nature of MRI, MR images are often used to segment and classify brain tumors. However, limited by the inaccuracy and inoperability of manual segmentation, it is very necessary to have a complete and comprehensive automatic brain tumor segmentation and classification algorithm technology. This article discusses the benefits, drawbacks, and areas of application of several traditional algorithms as well as more modern, improved, and more advanced algorithms. Segmentation methods and classification methods can be used to classify these techniques. Convolutional neural networks (CNN), Support vector machines, and Transformers are examples of classification methods. Random forests, decision trees, and improved U-Net algorithms are examples of segmentation methods. To discuss the capability of classification and segmentation, there are three sections in the area used for segmenting brain tumors with three types, including Tumor Core, Enhance Tumor, and Whole Tumor, which could be abbreviated as TC, ET, and WT. Through the comparative analysis of these methods, useful insights for future research are provided.