جستجوی مقالات مرتبط با کلیدواژه « Machine Learning » در نشریات گروه « پزشکی »

Recently, the medical profession has seen an accelerated integration of devices equipped with artificial intelligence (AI) technology, thanks to significant advancements in this area. Over 60 medical devices integrated with AI have already received approval from the Food and Drug Administration (FDA) in the United States. The widespread use of AI technology in medicine is seen as an unavoidable trend soon. AI technology is now being used in the area of cancer, particularly in radiology, for clinical use of medical equipment. It is anticipated that AI technology will become a crucial core technology in this industry. Precision medicine, which involves selecting the most suitable treatment for each patient based on extensive medical data like genome information, has gained global popularity. AI technology is anticipated to play a crucial role in extracting valuable data from vast medical datasets and applying it to medical care. Cancer is the second most prevalent global illness that relies on oncogenic mutations and non-mutated genes for its survival. The significant variability of tumors may result in varying curative results when using the same medications or surgical procedures in people with the same tumor. This highlights the need for more precise treatment approaches for tumors and personalized therapies tailored to individual patients. We summarize current and noteworthy AI advances in cancer research in this report. We also discuss AI’s limitations, challenges, and potential effects on cancer therapy. We also explored AI in omics, pathology, and medical imaging.

Exposure to air pollution is a major health problem worldwide. This study aimed to investigate the effect of the level of air pollutants and meteorological parameters with their related lag time on the transmission and severity of coronavirus disease 19 (COVID-19) using machine learning (ML) techniques in Shiraz, Iran.

Study Design:

 An ecological study.

In this ecological research, three main ML techniques, including decision trees, random forest, and extreme gradient boosting (XGBoost), have been applied to correlate meteorological parameters and air pollutants with infection transmission, hospitalization, and death due to COVID-19 from 1 October 2020 to 1 March 2022. These parameters and pollutants included particulate matter (PM2), sulfur dioxide (SO2 ), nitrogen dioxide (NO2 ), nitric oxide (NO), ozone (O3 ), carbon monoxide (CO), temperature (T), relative humidity (RH), dew point (DP), air pressure (AP), and wind speed (WS).

Based on the three ML techniques, NO2 (lag 5 day), CO (lag 4), and T (lag 25) were the most important environmental features affecting the spread of COVID-19 infection. In addition, the most important features contributing to hospitalization due to COVID-19 included RH (lag 28), T (lag 11), and O3 (lag 10). After adjusting for the number of infections, the most important features affecting the number of deaths caused by COVID-19 were NO2 (lag 20), O3 (lag 22), and NO (lag 23).

Our findings suggested that epidemics caused by COVID-19 and (possibly) similarly viral transmitted infections, including flu, air pollutants, and meteorological parameters, can be used to predict their burden on the community and health system. In addition, meteorological and air quality data should be included in preventive measures.

 Modeling with methods based on machine learning (ML) and artificial intelligence can help understand the complex relationships between ergonomic risk factors and employee health. The aim of this study was to use ML methods to estimate the effect of individual factors, ergonomic interventions, quality of work life (QWL), and productivity on work-related musculoskeletal disorders (WMSDs) in the neck area of office workers.

Study Design: 

A quasi-randomized control trial.

 To measure the impact of interventions, modeling with the ML method was performed on the data of a quasi-randomized control trial. The data included the information of 311 office workers (aged 32.04±5.34). Method neighborhood component analysis (NCA) was used to measure the effect of factors affecting WMSDs, and then support vector machines (SVMs) and decision tree algorithms were utilized to classify the decrease or increase of disorders.

 Three classified models were designed according to the follow-up times of the field study, with accuracies of 86.5%, 80.3%, and 69%, respectively. These models could estimate most influencer factors with acceptable sensitivity. The main factors included age, body mass index, interventions, QWL, some subscales, and several psychological factors. Models predicted that relative absenteeism and presenteeism were not related to the outputs.

 In this study, the focus was on disorders in the neck, and the obtained models revealed that individual and management interventions can be the main factors in reducing WMSDs in the neck. Modeling with ML methods can create a new understanding of the relationships between variables affecting WMSDs.

Detection of axillary metastases in breast cancer is critical for treatment options and prognosis. The aim of this study is to investigate the value of radiomic features obtained from short tau inversion recovery (STIR) sequences in magnetic resonance imaging (MRI) of primary tumor in breast cancer in predicting axillary lymph node metastasis (ALNM).

Lesions of 165 patients with a mean age of 51.12 ±11 (range 28-82) with newly diagnosed invasive breast cancer who underwent breast MRI before treatment were manually segmented from STIR sequences in the 3D Slicer program in all sections. Machine learning (ML) analysis was performed using the extracted 851 features Python 3.11, Pycaret library program. Datasets were randomly divided into train (123, 80%) and independent test (63, 20%) datasets. The performances of ML algorithms were compared with area under curve (AUC), accuracy, recall, presicion and F1 scores.

Accuracy and AUC in the training set were in the range of 57 %-86 % and 0.50-0.95, respectively. The best model in the training set was the catBoost classifier with an AUC of 0.95 and 84% accuracy. The AUC, accuracy, recall, precision values and F1 score of the CatBoost classifier on the test set were 0.92, 84 %, 89 %, 85 %, 86 %, respectively.

Radiomic features obtained from primary tumors on STIR sequences have the potential to predict ALNM in invasive breast cancer.

This study explores a novel approach to detecting arousal levels through the analysis of electroencephalography (EEG) signals. Leveraging the Faller database with data from 18 healthy participants, we employ a 64‑channel EEG system.

The approach we employ entails the extraction of ten frequency characteristics from every channel, culminating in a feature vector of 640 dimensions for each signal instance. To enhance classification accuracy, we employ a genetic algorithm for feature selection, treating it as a multiobjective optimization task. The approach utilizes fast bit hopping for efficiency, overcoming traditional bit‑string limitations. A hybrid operator expedites algorithm convergence, and a solution selection strategy identifies the most suitable feature subset.

Experimental results demonstrate the method’s effectiveness in detecting arousal levels across diverse states, with improvements in accuracy, sensitivity, and specificity. In scenario one, the proposed method achieves an average accuracy, sensitivity, and specificity of 93.11%, 98.37%, and 99.14%, respectively. In scenario two, the averages stand at 81.35%, 88.65%, and 84.64%.

The obtained results indicate that the proposed method has a high capability of detecting arousal levels in different scenarios. In addition, the advantage of employing the proposed feature reduction method has been demonstrated.

Occupational hearing loss of workers exposed to impulse noise and workers exposed to steady noise for a long time may have different clinical characteristics.

As of May 2019, all 92 servicemen working in a weapon experimental field exposed to impulse noise for over 1 year were collected as the impulse noise group. As of Dec 2019, all 78 servicemen working in an engine working experimental field exposed to steady noise for over 1 year were collected as the steady noise group. The propensity score matching (PSM) model was used to eliminate the imbalance of age and working time between the two groups of subjects. After propensity score matching, 51 subjects in each group were finally included in the study. The machine learning model is constructed according to pure tone auditory threshold, and the performance of the machine learning model is evaluated by accuracy, sensitivity, specificity, and AUC.

Subjects in the impulse noise group and the steady noise group had significant hearing loss at high frequencies. The hearing of the steady noise group was worse than that of the impulse noise group at speech frequency especially at the frequency of 1 kHz. Among machine learning models, XGBoost has the best prediction and classification performance.

The pure tone auditory threshold of subjects in both groups decreased and at high frequency. The hearing of the steady noise group at 1 kHz was significantly worse than that of the impulse noise group. XGBoost is the best model to predict the classification of our two groups. Our research can guide the prevention of damage caused by different types of noises.

Artificial Intelligence (AI), which mimics the human brain structure and operation, simulates intelligence. The aim of Machine Learning (ML), which is a branch of artificial intelligence, is to create models by analyzing data. Another type of artificial intelligence, Deep Learning (DL), depicts geometric changes using several layers of model representations. Since DL broke the computational analysis record, AI has advanced in many areas.

Contrary to the widespread use of conventional ML methodologies, there is still a need to promote the use and popularity of DL for pharmaceutical research and development. Drug discovery and design have been enhanced by ML and DL in major research projects. To fully realize its potential, drug design must overcome many challenges and issues. Various aspects of medication design must be considered to successfully address these concerns and challenges. This review article explains DL's significance both in technological breakthroughs and in effective medications.

There are numerous barriers and substantial challenges associated with drug design associated with DL architectures and key application domains. The article discusses several elements of medication development that have been influenced by existing research. Two widely used and efficient Neural Network (NN) designs are discussed in this article: Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs).

It is described how these tools can be utilized to design and discover small molecules for drug discovery. They are also given an overview of the history of DL approaches, as well as a discussion of some of their drawbacks.

As an artificial intelligence (AI) branch, machine learning has pioneering applications in public health, ranging from disease diagnosis to epidemic prediction. Machine learning (ML) is a strategic lever to improve care services’ access, quality, and efficiency and create health systems based on learning and value. In the following, we mention only a part of ML assistance in public health.

This study was carried out with the aim of determining weather parameters and air pollutants affecting seasonal changes of particulate matter of less than 10 microns (PM10) in Yazd city using Random Forest (RF) and extreme gradient boosting (Xgboost) models.

The required data was obtained from 2018 to 2022. Levene’s test was applied to investigate the significant difference in the variance of PM10 values in 4 different seasons, and Boruta algorithm was used to select the best predictive variables. RF and Xgboost models were trained using two-thirds of the input data and were tested using the remaining data set. Their performance was evaluated based on R2, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Nash–Sutcliffe Model Efficiency Coefficient (NSE).

The RF showed a higher performance in predicting PM10 in all the study seasons (R2  > 0.85; RMSE < 22). The contribution of dust concentration and relative humidity in spring PM10 changes was more than other variables. For summer, wind direction and ozone were identified as the most important variables affecting PM10 concentration. In the autumn and winter, air pollutants and dust concentration had the greatest effect on PM10, respectively.

RF model could explain more than 85% of PM10 seasonal variability in Yazd city. It is recommended to use the model to predict the changes of this air pollutant in other regions with similar climatic and environmental conditions. The results can also be useful for providing suitable solutions to reduce PM10 pollution hazards in Yazd city.

Artificial intelligence (AI) and its techniques are a rapidly growing field and are being used in various fields, including healthcare and many others. Medical entomology is one of the important sectors in health care. Diseases transmitted through carriers impose a great economic and social burden on the health of society. Mosquito-borne diseases pose major challenges to human health, affecting more than 600 million people and killing more than 1 million people each year. In the current study, we reviewed more than 30 papers in PubMed and Google Scholar that dealt with the application of artificial intelligence techniques in medical entomology. Articles were classified based on the use of AI and its techniques in this field and show that this new tool can play an important role in predicting the risk of contracting vector-borne diseases and the accurate monitoring of insect vector species.

Gliomas, as Central Nervous System (CNS) tumors, are greatly common with 80% of malignancy. Treatment methods for gliomas, such as surgery, radiation therapy, and chemotherapy depend on the grade, size, location, and the patient’s age.
This study aimed to quantify glioma based on the radiomics analysis and classify its grade into High-grade Glioma (HGG) or Low-grade Glioma (LGG) by various machine-learning methods using contrast-enhanced brain Computerized Tomography (CT) scans.
This retrospective study involved acquiring and segmenting data, selecting and extracting features, classifying, analyzing, and evaluating classifiers. The study included a total of 62 patients (31 with LGG and 31 with HGG). The tumors were segmented by an experienced CT-scan technologist with 3D slicer software. A total of 14 shape features, 18 histogram-based features, and 75 texture-based features were computed. The Area Under the Curve (AUC) and Receiver Operating Characteristic Curve (ROC) were used to evaluate and compare classification models.
A total of 13 out of 107 features were selected to differentiate between LGGs and HGGs and to perform various classifier algorithms with different cross-validations. The best classifier algorithm was linear-discriminant with 93.5% accuracy, 96.77% sensitivity, 90.3% specificity, and 0.98% AUC in the differentiation of LGGs and HGGs. 
The proposed method can identify LGG and HGG with 93.5% accuracy, 96.77% sensitivity, 90.3% specificity, and 0.98% AUC, leading to the best treatment for glioma patients by using CT scans based on radiomics analysis.

This study aims to diagnose the severity of important pathological indices, i.e., fibrosis, steatosis, lobular inflammation, and ballooning from the pathological images of the liver tissue based on extracted features by radiomics methods.

This research uses the pathological images obtained from liver tissue samples for 258 laboratory mice. After preprocessing the images and data augmentation, a collection of texture feature sets extracted by gray-level-based algorithms, including Global, Gray-level Co-Occurrence Matrix (GLCM), Gray-level Run length Matrix (GLRLM), Gray-level Size Zone Matrix (GLSZM), and Neighboring Gray Tone Difference Matrix (NGTDM) algorithms. Then, advanced methods of classification, namely Support Vector Machine (SVM), Random Forest (RF), Quadratic Discriminant Analysis (QDA), K-Nearest Neighbors (KNN), Logistic Regression (LR), Naïve Bayes (NB), and Multi-layer Perceptrons (MLP) are employed. This procedure is provided separately for each of the four indices of fibrosis level in 6 grading classes, steatosis in 5 grading classes, inflammation in 4 grading classes, and ballooning in 3 grading classes. For a comparison of the output of these algorithms, the accuracy value obtained from the evaluation data is presented for the performance of different methods.

The results showed that, compared to other methods, the Gaussian SVM algorithm provides a better response to the classification of the grading of liver disease among all the indices from the pathological images due to its structural features. This value of accuracy was calculated at 84.30% for fibrosis, 90.55% for steatosis, 81.11% for inflammation, and 95.98% for ballooning.

This fully automatic framework based on advanced radiomics algorithms and machine learning from pathological images can be very useful in clinical procedures and be considered as an assistant or a substitute for pathologists’ diagnoses.