Prediction of fetal weight at varying gestational age in the absence of ultrasound examination using ensemble learning

Prediction of fetal brain gestational age using multihead attention with Xception

Accurate gestational age (GA) prediction is crucial for monitoring fetal development and ensuring optimal prenatal care. Traditional methods often face challenges in terms of precision and prediction efficiency. In this context, leveraging modern deep learning (DL) techniques is a promising solution. This paper introduces a novel DL approach for GA prediction using fetal brain images obtained via magnetic resonance imaging (MRI), which combines the strength of the Xception pretrained model with a multihead attention (MHA) mechanism. The proposed model was trained on a diverse dataset comprising 52,900 fetal brain images from 741 patients. The images encompass a GA ranging from 19 to 39 weeks. These pretrained models served as feature extraction components during the training process. The extracted features were subsequently used as the inputs of different configurable MHAs, which produced GA predictions in days. The proposed model achieved promising results with 8 attention heads, 32 dimensionality of the key space and 32 dimensionality of the value space, with an R-squared (R2) value of 96.5 %, a mean absolute error (MAE) of 3.80 days, and a Pearson correlation coefficient (PCC) of 98.50 % for the test set. Additionally, the 5-fold cross-validation results reinforce the model's reliability, with an average R2 of 95.94 %, an MAE of 3.61 days, and a PCC of 98.02 %. The proposed model excels in different anatomical views, notably the axial and sagittal views. A comparative analysis of multiple planes and a single plane highlights the effectiveness of the proposed model against other state-of-the-art (SOTA) models reported in the literature. The proposed model could help clinicians accurately predict GA.

Assessment of supervised longitudinal learning methods: Insights from predicting low birth weight and very low birth weight using prenatal ultrasound measurements

BACKGROUND

This study aimed to assess the efficacy of various supervised longitudinal learning approaches, comparing traditional statistical models and machine learning algorithms for prediction with longitudinal data. The primary objectives were to evaluate the predictive performance of different supervised longitudinal learning methods for low birth weight (LBW) and very low birth weight (VLBW) based on prenatal ultrasound measurements. Additionally, the study sought to extract interpretable risk features for disease prediction.

METHODS

The evaluation involved benchmarking the performance of longitudinal models against conventional machine learning methods. Classification accuracy for LBW and VLBW at birth, as well as prediction accuracy for birth weight using prenatal sonographic ultrasound measurements, were assessed.

RESULTS

Among the learning approaches we investigated in this study, the longitudinal machine learning approach, specifically, the mixed effect random forest (MERF), delivered the overall best performance in predicting birthweights and classifying LBW/VLBW disease status.

CONCLUSION

The MERF combined the power of advanced machine learning algorithms to accommodate the inherent within-individual dependence in the observed data, delivering satisfactory performance in predicting the birthweight and classifying LBW/VLBW disease status. The study emphasized the importance of incorporating previous ultrasound measurements and considering correlations between repeated measurements for accurate prediction. The interpretable trees algorithm used for risk feature extraction proved reliable and applicable to other learning algorithms. These findings underscored the potential of longitudinal learning methods in improving birth weight prediction and highlighted the relevance of consistent risk features in line with established literature.

Machine learning approach for the prediction of macrosomia

Fetal macrosomia is associated with maternal and newborn complications due to incorrect fetal weight estimation or inappropriate choice of delivery models. The early screening and evaluation of macrosomia in the third trimester can improve delivery outcomes and reduce complications. However, traditional clinical and ultrasound examinations face difficulties in obtaining accurate fetal measurements during the third trimester of pregnancy. This study aims to develop a comprehensive predictive model for detecting macrosomia using machine learning (ML) algorithms. The accuracy of macrosomia prediction using logistic regression, k-nearest neighbors, support vector machine, random forest (RF), XGBoost, and LightGBM algorithms was explored. Each approach was trained and validated using data from 3244 pregnant women at a hospital in southern China. The information gain method was employed to identify deterministic features associated with the occurrence of macrosomia. The performance of six ML algorithms based on the recall and area under the curve evaluation metrics were compared. To develop an efficient prediction model, two sets of experiments based on ultrasound examination records within 1-7 days and 8-14 days prior to delivery were conducted. The ensemble model, comprising the RF, XGBoost, and LightGBM algorithms, showed encouraging results. For each experimental group, the proposed ensemble model outperformed other ML approaches and the traditional Hadlock formula. The experimental results indicate that, with the most risk-relevant features, the ML algorithms presented in this study can predict macrosomia and assist obstetricians in selecting more appropriate delivery models.

Fetal growth analysis from ultrasound videos based on different biometrics using optimal segmentation and hybrid classifier

Birth defects and their associated deaths, high health and financial costs of maternal care and associated morbidity are major contributors to infant mortality. If permitted by law, prenatal diagnosis allows for intrauterine care, more complicated hospital deliveries, and termination of pregnancy. During pregnancy, a set of measurements is commonly used to monitor the fetal health, including fetal head circumference, crown-rump length, abdominal circumference, and femur length. Because of the intricate interactions between the biological tissues and the US waves mother and fetus, analyzing fetal US images from a specialized perspective is difficult. Artifacts include acoustic shadows, speckle noise, motion blur, and missing borders. The fetus moves quickly, body structures close, and the weeks of pregnancy vary greatly. In this work, we propose a fetal growth analysis through US image of head circumference biometry using optimal segmentation and hybrid classifier. First, we introduce a hybrid whale with oppositional fruit fly optimization (WOFF) algorithm for optimal segmentation of segment fetal head which improves the detection accuracy. Next, an improved U-Net design is utilized for the hidden feature (head circumference biometry) extraction which extracts features from the segmented extraction. Then, we design a modified Boosting arithmetic optimization (MBAO) algorithm for feature optimization to selects optimal best features among multiple features for the reduction of data dimensionality issues. Furthermore, a hybrid deep learning technique called bi-directional LSTM with convolutional neural network (B-LSTM-CNN) for fetal growth analysis to compute the fetus growth and health. Finally, we validate our proposed method through the open benchmark datasets are HC18 (Ultrasound image) and oxford university research archive (ORA-data) (Ultrasound video frames). We compared the simulation results of our proposed algorithm with the existing state-of-art techniques in terms of various metrics.

Multimodal Early Birth Weight Prediction Using Multiple Kernel Learning

In this work, a novel multimodal learning approach for early prediction of birth weight is presented. Fetal weight is one of the most relevant indicators in the assessment of fetal health status. The aim is to predict early birth weight using multimodal maternal-fetal variables from the first trimester of gestation (Anthropometric data, as well as metrics obtained from Fetal Biometry, Doppler and Maternal Ultrasound). The proposed methodology starts with the optimal selection of a subset of multimodal features using an ensemble-based approach of feature selectors. Subsequently, the selected variables feed the nonparametric Multiple Kernel Learning regression algorithm. At this stage, a set of kernels is selected and weighted to maximize performance in birth weight prediction. The proposed methodology is validated and compared with other computational learning algorithms reported in the state of the art. The obtained results (absolute error of 234 g) suggest that the proposed methodology can be useful as a tool for the early evaluation and monitoring of fetal health status through indicators such as birth weight.

Placental MRI segmentation based on multi-receptive field and mixed attention separation mechanism

OBJECTIVE

To reduce the occurrence of massive bleeding during placental abruption in patients with placenta accrete, we established a medical imaging based on multi-receptive field and mixed attention separation mechanism (MRF-MAS) model to improve the accuracy of MRI placenta segmentation and provide a basis for subsequent placenta accreta.

METHODS

We propose a placenta MRI segmentation technology using the MRF-MAS framework to develop a medical image diagnostic technique. The model first uses the multi-receptive field feature structure to obtain multi-level information, and improves the expression of features at differing scales. Note that the hybrid attention mechanism combines channel attention and spatial attention, separates the input feature sets and computes the attention separately, and finally reorganizes the feature maps. To show that the model can improve the accuracy of segmenting the placenta, we adopt mean Intersection over Union (IoU), Dice similarity coefficient (Dice) and area under the receiver operating characteristic curve (AUC) with U-Net, Mask RCNN, Deeplab v3 for comparison.

RESULTS

The four models achieved different outcomes based on our placenta dataset, with our model IoU and Dice up to 0.8169 and 0.8992, which are 5.51% and 3.03% higher than the average of the three comparison models.

CONCLUSION

The model proposed by us is helpful to assist the imaging diagnosis and at the same time provides a quantitative reference for the precise treatment of placenta accreta, assists the Equationtion of the clinical operation plan of the physician, and promotes the precision medicine of placenta accreta.

Glenoid segmentation from computed tomography scans based on a 2-stage deep learning model for glenoid bone loss evaluation

BACKGROUND

The best-fitting circle drawn by computed tomography (CT) reconstruction of the en face view of the glenoid bone to measure the bone defect is widely used in clinical application. However, there are still some limitations in practical application, which can prevent the achievement of accurate measurements. This study aimed to accurately and automatically segment the glenoid from CT scans based on a 2-stage deep learning model and to quantitatively measure the glenoid bone defect.

MATERIALS AND METHODS

Patients who were referred to our institution between June 2018 and February 2022 were retrospectively reviewed. The dislocation group consisted of 237 patients with a history of ≥2 unilateral shoulder dislocations within 2 years. The control group consisted of 248 individuals with no history of shoulder dislocation, shoulder developmental deformity, or other disease that may lead to abnormal morphology of the glenoid. All patients underwent CT examination with a 1-mm slice thickness and a 1-mm increment, including complete imaging of the bilateral glenoid. A residual neural network (ResNet) location model and a U-Net bone segmentation model were constructed to develop an automated segmentation model for the glenoid from CT scans. The data set was randomly divided into training (201 of 248) and test (47 of 248) data sets of control-group data and training (190 of 237) and test (47 of 237) data sets of dislocation-group data. The accuracy of the stage 1 (glenoid location) model, the mean intersection-over-union value of the stage 2 (glenoid segmentation) model, and the glenoid volume error were used to assess the performance of the model. The R2 value and Lin concordance correlation coefficient were used to assess the correlation between the prediction and the gold standard.

RESULTS

A total of 73,805 images were obtained after the labeling process, and each image was composed of CT images of the glenoid and its corresponding mask. The average overall accuracy of stage 1 was 99.28%; the average mean intersection-over-union value of stage 2 was 0.96. The average glenoid volume error between the predicted and true values was 9.33%. The R2 values of the predicted and true values of glenoid volume and glenoid bone loss (GBL) were 0.87 and 0.91, respectively. The Lin concordance correlation coefficient value of the predicted and true values of glenoid volume and GBL were 0.93 and 0.95, respectively.

CONCLUSION

The 2-stage model in this study showed a good performance in glenoid bone segmentation from CT scans and could quantitatively measure GBL, providing a data reference for subsequent clinical treatment.

BabyNet＋＋: Fetal birth weight prediction using biometry multimodal data acquired less than 24 hours before delivery

Accurate prediction of fetal weight at birth is essential for effective perinatal care, particularly in the context of antenatal management, which involves determining the timing and mode of delivery. The current standard of care involves performing a prenatal ultrasound 24 hours prior to delivery. However, this task presents challenges as it requires acquiring high-quality images, which becomes difficult during advanced pregnancy due to the lack of amniotic fluid. In this paper, we present a novel method that automatically predicts fetal birth weight by using fetal ultrasound video scans and clinical data. Our proposed method is based on a Transformer-based approach that combines a Residual Transformer Module with a Dynamic Affine Feature Map Transform. This method leverages tabular clinical data to evaluate 2D+t spatio-temporal features in fetal ultrasound video scans. Development and evaluation were carried out on a clinical set comprising 582 2D fetal ultrasound videos and clinical records of pregnancies from 194 patients performed less than 24 hours before delivery. Our results show that our method outperforms several state-of-the-art automatic methods and estimates fetal birth weight with an accuracy comparable to human experts. Hence, automatic measurements obtained by our method can reduce the risk of errors inherent in manual measurements. Observer studies suggest that our approach may be used as an aid for less experienced clinicians to predict fetal birth weight before delivery, optimizing perinatal care regardless of the available expertise.

Deep learning for estimation of fetal weight throughout the pregnancy from fetal abdominal ultrasound

BACKGROUND

Fetal weight is currently estimated from fetal biometry parameters using heuristic mathematical formulas. Fetal biometry requires measurements of the fetal head, abdomen, and femur. However, this examination is prone to inter- and intraobserver variability because of factors, such as the experience of the operator, image quality, maternal characteristics, or fetal movements. Our study tested the hypothesis that a deep learning method can estimate fetal weight based on a video scan of the fetal abdomen and gestational age with similar performance to the full biometry-based estimations provided by clinical experts.

OBJECTIVE

This study aimed to develop and test a deep learning method to automatically estimate fetal weight from fetal abdominal ultrasound video scans.

STUDY DESIGN

A dataset of 900 routine fetal ultrasound examinations was used. Among those examinations, 800 retrospective ultrasound video scans of the fetal abdomen from 700 pregnant women between 15 6/7 and 41 0/7 weeks of gestation were used to train the deep learning model. After the training phase, the model was evaluated on an external prospectively acquired test set of 100 scans from 100 pregnant women between 16 2/7 and 38 0/7 weeks of gestation. The deep learning model was trained to directly estimate fetal weight from ultrasound video scans of the fetal abdomen. The deep learning estimations were compared with manual measurements on the test set made by 6 human readers with varying levels of expertise. Human readers used standard 3 measurements made on the standard planes of the head, abdomen, and femur and heuristic formula to estimate fetal weight. The Bland-Altman analysis, mean absolute percentage error, and intraclass correlation coefficient were used to evaluate the performance and robustness of the deep learning method and were compared with human readers.

RESULTS

Bland-Altman analysis did not show systematic deviations between readers and deep learning. The mean and standard deviation of the mean absolute percentage error between 6 human readers and the deep learning approach was 3.75%±2.00%. Excluding junior readers (residents), the mean absolute percentage error between 4 experts and the deep learning approach was 2.59%±1.11%. The intraclass correlation coefficients reflected excellent reliability and varied between 0.9761 and 0.9865.

CONCLUSION

This study reports the use of deep learning to estimate fetal weight using only ultrasound video of the fetal abdomen from fetal biometry scans. Our experiments demonstrated similar performance of human measurements and deep learning on prospectively acquired test data. Deep learning is a promising approach to directly estimate fetal weight using ultrasound video scans of the fetal abdomen.

Machine learning for predicting diabetes risk in western China adults

OBJECTIVE

Diabetes mellitus is a global epidemic disease. Long-time exposure of patients to hyperglycemia can lead to various type of chronic tissue damage. Early diagnosis of and screening for diabetes are crucial to population health.

METHODS

We collected the national physical examination data in Xinjiang, China, in 2020 (a total of more than 4 million people). Three types of physical examination indices were analyzed: questionnaire, routine physical examination and laboratory values. Integrated learning, deep learning and logistic regression methods were used to establish a risk model for type-2 diabetes mellitus. In addition, to improve the convenience and flexibility of the model, a diabetes risk score card was established based on logistic regression to assess the risk of the population.

RESULTS

An XGBoost-based risk prediction model outperformed the other five risk assessment algorithms. The AUC of the model was 0.9122. Based on the feature importance ranking map, we found that hypertension, fasting blood glucose, age, coronary heart disease, ethnicity, parental diabetes mellitus, triglycerides, waist circumference, total cholesterol, and body mass index were the most important features of the risk prediction model for type-2 diabetes.

CONCLUSIONS

This study established a diabetes risk assessment model based on multiple ethnicities, a large sample and many indices, and classified the diabetes risk of the population, thus providing a new forecast tool for the screening of patients and providing information on diabetes prevention for healthy populations.

Machine learning and disease prediction in obstetrics

Machine learning technologies and translation of artificial intelligence tools to enhance the patient experience are changing obstetric and maternity care. An increasing number of predictive tools have been developed with data sourced from electronic health records, diagnostic imaging and digital devices. In this review, we explore the latest tools of machine learning, the algorithms to establish prediction models and the challenges to assess fetal well-being, predict and diagnose obstetric diseases such as gestational diabetes, pre-eclampsia, preterm birth and fetal growth restriction. We discuss the rapid growth of machine learning approaches and intelligent tools for automated diagnostic imaging of fetal anomalies and to asses fetoplacental and cervix function using ultrasound and magnetic resonance imaging. In prenatal diagnosis, we discuss intelligent tools for magnetic resonance imaging sequencing of the fetus, placenta and cervix to reduce the risk of preterm birth. Finally, the use of machine learning to improve safety standards in intrapartum care and early detection of complications will be discussed. The demand for technologies to enhance diagnosis and treatment in obstetrics and maternity should improve frameworks for patient safety and enhance clinical practice.

RSU-Net: U-net based on residual and self-attention mechanism in the segmentation of cardiac magnetic resonance images

BACKGROUND

Automated segmentation techniques for cardiac magnetic resonance imaging (MRI) are beneficial for evaluating cardiac functional parameters in clinical diagnosis. However, due to the characteristics of unclear image boundaries and anisotropic resolution anisotropy produced by cardiac magnetic resonance imaging technology, most of the existing methods still have the problems of intra-class uncertainty and inter-class uncertainty. However, due to the irregularity of the anatomical shape of the heart and the inhomogeneity of tissue density, the boundaries of its anatomical structures become uncertain and discontinuous. Therefore, fast and accurate segmentation of cardiac tissue remains a challenging problem in medical image processing.

METHODOLOGY

We collected cardiac MRI data from 195 patients as training set and 35patients from different medical centers as external validation set. Our research proposed a U-net network architecture with residual connections and a self-attentive mechanism (Residual Self-Attention U-net, RSU-Net). The network relies on the classic U-net network, adopts the U-shaped symmetric architecture of the encoding and decoding mode, improves the convolution module in the network, introduces skip connections, and improves the network's capacity for feature extraction. Then for solving locality defects of ordinary convolutional networks. To achieve a global receptive field, a self-attention mechanism is introduced at the bottom of the model. The loss function employs a combination of Cross Entropy Loss and Dice Loss to jointly guide network training, resulting in more stable network training.

RESULTS

In our study, we employ the Hausdorff distance (HD) and the Dice similarity coefficient (DSC) as metrics for assessing segmentation outcomes. Comparsion was made with the segmentation frameworks of other papers, and the comparison results prove that our RSU-Net network performs better and can make accurate segmentation of the heart. New ideas for scientific research.

CONCLUSION

Our proposed RSU-Net network combines the advantages of residual connections and self-attention. This paper uses the residual links to facilitate the training of the network. In this paper, a self-attention mechanism is introduced, and a bottom self-attention block (BSA Block) is used to aggregate global information. Self-attention aggregates global information, and has achieved good segmentation results on the cardiac segmentation dataset. It facilitates the diagnosis of cardiovascular patients in the future.

Mining attention distribution paradigm: Discover gaze patterns and their association rules behind the visual image

BACKGROUND AND OBJECTIVE

Attention allocation reflects the way of humans filtering and organizing the information. On one hand, different task scenarios seriously affect human's rule of attention distribution, on the other hand, visual attention reflecting the cognitive and psychological process. Most of the previous studies on visual attention allocation are based on cognitive models, predicted models, or statistical analysis of eye movement data or visual images, however, these methods are inadequate to provide an inside view of gaze behavior to reveal the attention distribution pattern within scenario context. Moreover, they seldom study the association rules of these patterns. Therefore, we adopted the big data mining approach to discover the paradigm of visual attention distribution.

METHODS

We applied the data mining method to extract the gaze patterns to discover the regularities of attention distribution behavior within the scenario context. The proposed method consists of three components, tasks scenario segmented and clustered, gaze pattern mining, and association rule of frequent pattern mining.

RESULTS

The proposed approach is tested on the operation platform. The complex operation task is simultaneously segmented and clustered with the TICC-based method and evaluated by the BCI index. The operator's eye movement frequent patterns and their association rule are discovered. The results demonstrate that our method can associate the eye-tracking data with the task-oriented scene data.

DISCUSSION

The proposed method provides the benefits of being able to explicitly express and quantitatively analyze people's visual attention patterns. The proposed method can not only be applied in the field of aerospace medicine and aviation psychology, but also can likely be applied to computer-aided diagnosis and follow-up tool for neurological disease and cognitive impairment related disease, such as ADHD (Attention Deficit Hyperactivity Disorder), neglect syndrome, social attention differences in ASD (Autism spectrum disorder).

Stroke risk prediction by color Doppler ultrasound of carotid artery-based deep learning using Inception V3 and VGG-16

Purpose

This study aims to automatically classify color Doppler images into two categories for stroke risk prediction based on the carotid plaque. The first category is high-risk carotid vulnerable plaque, and the second is stable carotid plaque.

Method

In this research study, we used a deep learning framework based on transfer learning to classify color Doppler images into two categories: one is high-risk carotid vulnerable plaque, and the other is stable carotid plaque. The data were collected from the Second Affiliated Hospital of Fujian Medical University, including stable and vulnerable cases. A total of 87 patients with risk factors for atherosclerosis in our hospital were selected. We used 230 color Doppler ultrasound images for each category and further divided those into the training set and test set in a ratio of 70 and 30%, respectively. We have implemented Inception V3 and VGG-16 pre-trained models for this classification task.

Results

Using the proposed framework, we implemented two transfer deep learning models: Inception V3 and VGG-16. We achieved the highest accuracy of 93.81% by using fine-tuned and adjusted hyperparameters according to our classification problem.

Conclusion

In this research, we classified color Doppler ultrasound images into high-risk carotid vulnerable and stable carotid plaques. We fine-tuned pre-trained deep learning models to classify color Doppler ultrasound images according to our dataset. Our suggested framework helps prevent incorrect diagnoses caused by low image quality and individual experience, among other factors.

Bone age recognition based on mask R-CNN using xception regression model

Background and Objective: Bone age detection plays an important role in medical care, sports, judicial expertise and other fields. Traditional bone age identification and detection is according to manual interpretation of X-ray images of hand bone by doctors. This method is subjective and requires experience, and has certain errors. Computer-aided detection can effectually enhance the validity of medical diagnosis, especially with the fast development of machine learning and neural network, the method of bone age recognition using machine learning has gradually become the focus of research, which has the advantages of simple data pretreatment, good robustness and high recognition accuracy. Methods: In this paper, the hand bone segmentation network based on Mask R-CNN was proposed to segment the hand bone area, and the segmented hand bone region was directly input into the regression network for bone age evaluation. The regression network is using an enhancd network Xception of InceptionV3. After the output of Xception, the convolutional block attention module is connected to refine the feature mapping from channel and space to obtain more effective features. Results: According to the experimental results, the hand bone segmentation network model based on Mask R-CNN can segment the hand bone region and eliminate the interference of redundant background information. The average Dice coefficient on the verification set is 0.976. The mean absolute error of predicting bone age on our data set was only 4.97 months, which exceeded the accuracy of most other bone age assessment methods. Conclusion: Experiments show that the accuracy of bone age assessment can be enhancd by using the Mask R-CNN-based hand bone segmentation network and the Xception bone age regression network to form a model, which can be well applied to actual clinical bone age assessment.

A deep learning based framework for the classification of multi- class capsule gastroscope image in gastroenterologic diagnosis

Purpose: The purpose of this paper is to develop a method to automatic classify capsule gastroscope image into three categories to prevent high-risk factors for carcinogenesis, such as atrophic gastritis (AG). The purpose of this research work is to develop a deep learning framework based on transfer learning to classify capsule gastroscope image into three categories: normal gastroscopic image, chronic erosive gastritis images, and ulcer gastric image. Method: In this research work, we proposed deep learning framework based on transfer learning to classify capsule gastroscope image into three categories: normal gastroscopic image, chronic erosive gastritis images, and ulcer gastric image. We used VGG- 16, ResNet-50, and Inception V3 pre-trained models, fine-tuned them and adjust hyperparameters according to our classification problem. Results: A dataset containing 380 images was collected for each capsule gastroscope image category, and divided into training set and test set in a ratio of 70%, and 30% respectively, and then based on the dataset, three methods, including as VGG- 16, ResNet-50, and Inception v3 are used. We achieved highest accuracy of 94.80% by using VGG- 16 to diagnose and classify capsule gastroscopic images into three categories: normal gastroscopic image, chronic erosive gastritis images, and ulcer gastric image. Our proposed approach classified capsule gastroscope image with respectable specificity and accuracy. Conclusion: The primary technique and industry standard for diagnosing and treating numerous stomach problems is gastroscopy. Capsule gastroscope is a new screening tool for gastric diseases. However, a number of elements, including image quality of capsule endoscopy, the doctors' experience and fatigue, limit its effectiveness. Early identification is necessary for high-risk factors for carcinogenesis, such as atrophic gastritis (AG). Our suggested framework will help prevent incorrect diagnoses brought on by low image quality, individual experience, and inadequate gastroscopy inspection coverage, among other factors. As a result, the suggested approach will raise the standard of gastroscopy. Deep learning has great potential in gastritis image classification for assisting with achieving accurate diagnoses after endoscopic procedures.

High-resolution computed tomography diagnosis of pneumoconiosis complicated with pulmonary tuberculosis based on cascading deep supervision U-Net

OBJECTIVE

Pulmonary tuberculosis can promote pneumoconiosis deterioration, leading to higher mortality. This study aims to explore the diagnostic value of the cascading deep supervision U-Net (CSNet) model in pneumoconiosis complicated with pulmonary tuberculosis.

METHODS

A total of 162 patients with pneumoconiosis treated in our hospital were collected as the research objects. Patients were randomly divided into a training set (n = 113) and a test set (n = 49) in proportion (7:3). Based on the high-resolution computed tomography (HRCT), the traditional U-Net, supervision U-Net (SNet), and CSNet prediction models were constructed. Dice similarity coefficients, precision, recall, volumetric overlap error, and relative volume difference were used to evaluate the segmentation model. The area under the receiver operating characteristic curve (AUC) value represents the prediction efficiency of the model.

RESULTS

There were no statistically significant differences in gender, age, number of positive patients, and dust contact time between patients in the training set and test set (P > 0.05). The segmentation results of CSNet are better than the traditional U-Net model and the SNet model. The AUC value of the CSNet model was 0.947 (95% CI: 0.900∼0.994), which was higher than the traditional U-Net model.

CONCLUSION

The CSNet based on chest HRCT proposed in this study is superior to the traditional U-Net segmentation method in segmenting pneumoconiosis complicated with pulmonary tuberculosis. It has good prediction efficiency and can provide more clinical diagnostic value.

Transcription factor NFE2L3 promotes the proliferation of esophageal squamous cell carcinoma cells and causes radiotherapy resistance by regulating IL-6

OBJECTIVE

To scrutinize the impact of overexpression and interference of NFE2L3 on radiosensitivity of esophageal squamous cell carcinoma cells (ESCC) and its downstream mechanism and to assess whether NFE2L3 expression alters in vivo radiosensitivity of ESCC by developing a subcutaneous tumor model in mice.

METHODS

Through RNA-Seq, we compared the differentially expressed genes between the ECA-109R cell line and its parent ECA-109 cell line. The differentially expressed genes were selected and verified by qRT-PCR. Transfection of ESCC cell lines with NFE2L3 inhibitor or mimic lentivirus constructs was done to study the activity of NFE2L3. To assess the effect of NFE2L3 on cellular growth and proliferation, clonogenic survival assay, EdU incorporation assay, and CCK-8 assay were done after irradiation. To probe how many irradiated DNA double-strand breaks were produced, the corresponding intensity of γ-H2AX foci were detected by immunofluorescence. Apoptotic cells were assayed by flow cytometry assay after irradiation; To investigate the downstream genes of NFE2L3, we knocked NFE2L3, and RNA-Seq was used to find out the downstream genes. qRT-PCR and western blot ensued to score associated protein profiles. The in vivo ESCC cell radiosensitivity was scrutinized by nude mouse xenograft models.

RESULTS

The differential genes between ECA-109R cells and its parent ECA-109 cells were compared by qRT-PCR to unveil a significant increase in NFE2L3 expression. Functional analysis indicated that NFE2L3 increased radioresistance in ESCC cells. Then, through high-throughput sequencing and bioinformatics analysis, IL-6 was found to be a hub gene that played a role downstream of NFE2L3 and was verified by qRT-PCR, western blot, and double luciferase reporter gene experiment. NFE2L3 could regulate ESCC cell radiosensitivity via the IL-6-STAT3 signaling pathway, and downregulation of IL-6 expression could reverse the effects of highly expressed NFE2L3. In vivo tumor xenograft experiments confirmed that NFE2L3 affects the sensitivity to radiation therapy.

CONCLUSION

NFE2L3 can affect the radiosensitivity of ESCC cells through IL-6 transcription and IL-6/STAT3 signaling pathway. This makes NFE2L3 a putative target to regulate ESCC cell radiosensitivity.

Differentiation and prediction of pneumoconiosis stage by computed tomography texture analysis based on U-Net neural network

BACKGROUND AND OBJECTIVE

The progressive worsening of pneumoconiosis will ensue a hazardous physical condition in patients. This study details the differential diagnosis of the pneumoconiosis stage, by employing computed tomography (CT) texture analysis, based on U-Net neural network.

METHODS

The pneumoconiosis location from 92 patients at various stages was extracted by U-Net neural network. Mazda software was employed to analyze the texture features. Three dimensionality reduction methods set the best texture parameters. We applied four methods of the B11 module to analyze the selected texture parameters and calculate the misclassified rate (MCR). Finally, the receiver operating characteristic curve (ROC) of the texture parameters was analyzed, and the texture parameters with diagnostic efficiency were evaluated by calculating the area under curve (AUC).

RESULTS

The original film was processed by Gaussian and Laplace filters for a better display of the segmented area of pneumoconiosis in all stages. The MCR value obtained by the NDA analysis method under the MI dimension reduction method was the lowest, at 10.87%. In the filtered texture feature parameters, the best AUC was 0.821.

CONCLUSIONS

CT texture analysis based on the U-Net neural network can be used to identify the staging of pneumoconiosis.

Classification of lungs infected COVID-19 images based on inception-ResNet

OBJECTIVE

Nowadays, COVID-19 is spreading rapidly worldwide, and seriously threatening lives . From the perspective of security and economy, the effective control of COVID-19 has a profound impact on the entire society. An effective strategy is to diagnose earlier to prevent the spread of the disease and prompt treatment of severe cases to improve the chance of survival.

METHODS

The method of this paper is as follows: Firstly, the collected data set is processed by chest film image processing, and the bone removal process is carried out in the rib subtraction module. Then, the set preprocessing method performed histogram equalization, sharpening, and other preprocessing operations on the chest film. Finally, shallow and high-level feature mapping through the backbone network extracts the processed chest radiographs. We implement the self-attention mechanism in Inception-Resnet, perform the standard classification, and identify chest radiograph diseases through the classifier to realize the auxiliary COVID-19 diagnosis process at the medical level, all in an effort to further enhance the classification performance of the convolutional neural network. Numerous computer simulations demonstrate that the Inception-Resnet convolutional neural network performs CT image categorization and enhancement with greater efficiency and flexibility than conventional segmentation techniques.

RESULTS

The experimental COVID-19 CT dataset obtained in this paper is the new data for CT scans and medical imaging of normal, early COVID-19 patients and severe COVID-19 patients from Jinyintan hospital. The experiment plots the relationship between model accuracy, model loss and epoch, using ACC, TPR, SPE, F1 score and G-mean to measure the image maps of patients with and without the disease. Statistical measurement values are obtained by Inception-Resnet are 88.23%, 83.45%, 89.72%, 95.53% and 88.74%. The experimental results show that Inception-Resnet plays a more effective role than other image classification methods in evaluation indicators, and the method has higher robustness, accuracy and intuitiveness.

CONCLUSION

With CT images in the clinical diagnosis of COVID-19 images being widely used and the number of applied samples continuously increasing, the method in this paper is expected to become an additional diagnostic tool that can effectively improve the diagnostic accuracy of clinical COVID-19 images.

Combining machine learning with radiomics features in predicting outcomes after mechanical thrombectomy in patients with acute ischemic stroke

BACKGROUND AND OBJECTIVE

Some patients with mechanical thrombectomy will have a poor prognosis. This study establishes a model for predicting the prognosis after mechanical thrombectomy in acute stroke based on diffusion-weighted imaging (DWI) omics characteristics.

METHODS

A total of 260 stroke patients receiving mechanical thrombectomy in our hospital were randomly divided into a training set (n = 182) and a test set (n = 78) in a 7:3 ratio. The regions of interest (ROI) of the imaging features of the DWI infarct area were extracted, and the minimum absolute contraction and selection operator regression model were used to screen the best radiomics features. A support vector machine classifier established the prediction model of the prognosis after mechanical thrombectomy of acute stroke based on the selected features. The prediction efficiency of the model was evaluated by the receiver operating characteristic (ROC) curve.

RESULTS

A total of 1936 radiomic features were extracted, and six features highly correlated with prognosis were screened after dimensionality reduction. Based on the DWI model, the ROC analysis showed that the area under the curve (AUC) for correct prediction in the training and test sets was 0.945 and 0.920, respectively.

CONCLUSION

The model based on the characteristics of radiomics and machine learning has high predictive efficiency for the prognosis of acute stroke after mechanical thrombectomy, which can be used to guide personalized clinical treatment.

Deep learning-based medical image segmentation of the aorta using XR-MSF-U-Net

PURPOSE

This paper proposes a CT images and MRI segmentation technology of cardiac aorta based on XR-MSF-U-Net model. The purpose of this method is to better analyze the patient's condition, reduce the misdiagnosis and mortality rate of cardiovascular disease in inhabitants, and effectively avoid the subjectivity and unrepeatability of manual segmentation of heart aorta, and reduce the workload of doctors.

METHOD

We implement the X ResNet (XR) convolution module to replace the different convolution kernels of each branch of two-layer convolution XR of common model U-Net, which can make the model extract more useful features more efficiently. Meanwhile, a plug and play attention module integrating multi-scale features Multi-scale features fusion module (MSF) is proposed, which integrates global local and spatial features of different receptive fields to enhance network details to achieve the goal of efficient segmentation of cardiac aorta through CT images and MRI.

RESULTS

The model is trained on common cardiac CT images and MRI data sets and tested on our collected data sets to verify the generalization ability of the model. The results show that the proposed XR-MSF-U-Net model achieves a good segmentation effect on CT images and MRI. In the CT data set, the XR-MSF-U-Net model improves 7.99% in key index DSC and reduces 11.01 mm in HD compared with the benchmark model U-Net, respectively. In the MRI data set, XR-MSF-U-Net model improves 10.19% and reduces 6.86 mm error in key index DSC and HD compared with benchmark model U-Net, respectively. And it is superior to similar models in segmentation effect, proving that this model has significant advantages.

CONCLUSION

This study provides new possibilities for the segmentation of aortic CT images and MRI, improves the accuracy and efficiency of diagnosis, and hopes to provide substantial help for the segmentation of aortic CT images and MRI.

Blind Image Inpainting with Mixture Noise Using ℓ 0 and Total Regularization

The blind image inpainting problem need to be handle when faced with a large number of images, especially medical images in medical health. For the proposed nonconvex sparse optimization model, a proximal based alternating direction method of multipliers (PADMM) method is designed to solve the problem. Firstly, ℓ 0 sparse regularization is imposed to the binary mask since the missing pixels are sparse in our experiments. Secondly, the total variation term is utilized to describe the underlying clean image. Finally, ℓ 2 regularization of the fidelity term is used to solve the given blind inpainting problem. Experiments show that this method has better performance than traditional method, and could deal with the blind image inpainting problem.

Magnetic resonance analysis of deep cerebral venous vasospasm after subarachnoid hemorrhage in rabbits

Objective

Arterial spasm is proved to be an inducer of cerebral ischemia and cerebral infarction, while when a venous spasm occurs, cerebral edema is seen to be caused by a disturbance in cerebral blood flow. However, it is unclear and unproven whether venous spasm occurs after subarachnoid hemorrhage (SAH). To provide the theoretical basis for treating cerebral vasospasm after SAH, magnetic resonance imaging (MRI) was employed to observe the changes in the diameter of deep cerebral veins in rabbits after SAH.

Methods

Fourteen New Zealand rabbits were randomly divided into the SAH group (n = 10) and the normal saline group (NS group, n = 4). Specifically, the SAH models were established by the ultrasound-guided double injections of blood into cisterna magna. Moreover, the MRI was performed to observe the changes in the diameter of deep cerebral veins (internal cerebral vein, basilar vein, and great cerebral vein) and basilar artery before modeling (0 d) and 1, 3, 5, 7, 9, and 11 d after modeling.

Results

In the SAH group, the diameter of the basilar artery showed no evident change on the 1st d. However, it became narrower obviously on the 3rd d and 5th d, and the stenosis degree was more than 30%. The diameter gradually relieved from 7th to 9th d, and finally returned to normal on the 11th d. Moreover, the diameter of the internal cerebral vein significantly narrowed on the 1st d, the stenosis degree of which was 19%; the stenosis then relieved slightly on the 3rd d (13%), reached the peak (34%) on the 5th d, and gradually relieved from 7th d to 11th d. Moreover, the stenosis degree of the basilar vein was 18% on the 1st d, 24% on the 3rd d, and reached the peak (34%) on the 5th d.

Conclusion

After SAH in rabbits, the cerebral vasospasm was seen to occur in the basilar artery, and likewise, spasmodic changes took place in the deep cerebral vein. Furthermore, the time regularity of spasmodic changes between the cerebral vein and basilar artery was of significant difference, indicating that the venous vasospasm resulted in active contraction.

Breast MRI Tumor Automatic Segmentation and Triple-Negative Breast Cancer Discrimination Algorithm Based on Deep Learning

Background

Breast cancer is a kind of cancer that starts in the epithelial tissue of the breast. Breast cancer has been on the rise in recent years, with a younger generation developing the disease. Magnetic resonance imaging (MRI) plays an important role in breast tumor detection and treatment planning in today's clinical practice. As manual segmentation grows more time-consuming and the observed topic becomes more diversified, automated segmentation becomes more appealing. Methodology. For MRI breast tumor segmentation, we propose a CNN-SVM network. The labels from the trained convolutional neural network are output using a support vector machine in this technique. During the testing phase, the convolutional neural network's labeled output, as well as the test grayscale picture, is passed to the SVM classifier for accurate segmentation.

Results

We tested on the collected breast tumor dataset and found that our proposed combined CNN-SVM network achieved 0.93, 0.95, and 0.92 on DSC coefficient, PPV, and sensitivity index, respectively. We also compare with the segmentation frameworks of other papers, and the comparison results prove that our CNN-SVM network performs better and can accurately segment breast tumors.

Conclusion

Our proposed CNN-SVM combined network achieves good segmentation results on the breast tumor dataset. The method can adapt to the differences in breast tumors and segment breast tumors accurately and efficiently. It is of great significance for identifying triple-negative breast cancer in the future.

Image Recognition of Pediatric Pneumonia Based on Fusion of Texture Features and Depth Features

Pneumonia is one of the diseases that seriously endangers human health, and it is also the leading cause of death of children under the age of five in China. The most commonly used imaging examination method for radiologists is mainly based on chest X-ray images. Still, imaging errors often result during imaging examinations due to objective factors such as visual fatigue and lack of experience. Therefore, this paper proposes a feature fusion model, FC-VGG, based on the fusion of texture features (local binary pattern LBP and directional gradient histogram HOG) and depth features. The model improves model performance by adding detailed information in texture features to the convolutional neural network while making the model more suitable for clinical use. We input the X-ray image with texture features into the modified VGG16 model, C-VGG, and then add the Add fusion method to C-VGG for feature fusion so that FC-VGG is obtained, so FC-VGG has texture features detailed information and abstract information of deep features. Through experiments, our model has achieved 92.19% accuracy in recognizing children's pneumonia images, 93.44% average precision, 92.19% average recall, and 92.81% average F1 coefficient, and the model performance exceeds existing deep learning models and traditional feature recognition algorithms.

Risk Factors and Prediction Models for Nonalcoholic Fatty Liver Disease Based on Random Forest

Objective

To establish a risk prediction model of nonalcoholic fatty liver disease (NAFLD) and provide management strategies for preventing this disease.

Methods

A total of 200 inpatients and physical examinees were collected from the Department of Gastroenterology and Endocrinology and Physical Examination Center. The data of physical examination, laboratory examination, and abdominal ultrasound examination were collected. All subjects were randomly divided into a training set (70%) and a verification set (30%). A random forest (RF) prediction model is constructed to predict the occurrence risk of NAFLD. The receiver operating characteristic (ROC) curve is used to verify the prediction effect of the prediction models.

Results

The number of NAFLD patients was 44 out of 200 enrolled patients, and the cumulative incidence rate was 22%. The prediction models showed that BMI, TG, HDL-C, LDL-C, ALT, SUA, and MTTP mutations were independent influencing factors of NAFLD, all of which has statistical significance (P < 0.05). The area under curve (AUC) of logistic regression and the RF model was 0.940 (95% CI: 0.870~0.987) and 0.945 (95% CI: 0.899~0.994), respectively.

Conclusion

This study established a prediction model of NAFLD occurrence risk based on the RF, which has a good prediction value.

Convolutional Neural Network in Microsurgery Treatment of Spontaneous Intracerebral Hemorrhage

Objective

To explore the convolutional neural network (CNN) method in measuring hematoma volume-assisted microsurgery for spontaneous cerebral hemorrhage.

Methods

A total of 120 patients with spontaneous cerebral hemorrhage were selected and randomly divided into control and CNN groups with 60 patients in each group. Patients in the control group received traditional Tada formula to calculate hematoma volume and microsurgery. Convolutional neural network algorithm segmentation was used to measure hematoma volume, and microsurgery was performed in the CNN group. This article assessed neurological function, ability to live daily, complication rate, and prognosis.

Results

The incidence of postoperative complications in the CNN group (13.33%) was lower than the control group (43.33%). The neurological function and daily living ability in the CNN groups were recovered better. The incidence of poor prognosis in the CNN group (16.67%) was lower than the control group (30.00%).

Conclusion

Convolutional neural network measurement of hematoma volume to assist microsurgical treatment of spontaneous intracerebral hemorrhage patients is conducive to early recovery, reducing the damage to the patients' cerebral nerves.

Sensing and Artificial Intelligent Maternal-Infant Health Care Systems: A Review

Currently, information and communication technology (ICT) allows health institutions to reach disadvantaged groups in rural areas using sensing and artificial intelligence (AI) technologies. Applications of these technologies are even more essential for maternal and infant health, since maternal and infant health is vital for a healthy society. Over the last few years, researchers have delved into sensing and artificially intelligent healthcare systems for maternal and infant health. Sensors are exploited to gauge health parameters, and machine learning techniques are investigated to predict the health conditions of patients to assist medical practitioners. Since these healthcare systems deal with large amounts of data, significant development is also noted in the computing platforms. The relevant literature reports the potential impact of ICT-enabled systems for improving maternal and infant health. This article reviews wearable sensors and AI algorithms based on existing systems designed to predict the risk factors during and after pregnancy for both mothers and infants. This review covers sensors and AI algorithms used in these systems and analyzes each approach with its features, outcomes, and novel aspects in chronological order. It also includes discussion on datasets used and extends challenges as well as future work directions for researchers.

An accurate and explainable ensemble learning method for carotid plaque prediction in an asymptomatic population

BACKGROUND AND OBJECTIVE

The identification of carotid plaque, one of the most crucial tasks in stroke screening, is of great significance in the assessment of subclinical atherosclerosis and preventing the onset of stroke. However, traditional ultrasound examination is not prevalent or cost-effective for asymptomatic people, particularly low-income individuals in rural areas. Thus, it is necessary to develop an accurate and explainable model for early identification of the risk of plaque prevalence that can help in the primary prevention of stroke.

METHODS

We developed an ensemble learning method to predict the occurrence of carotid plaques. A dataset comprising 1440 subjects (50% with plaques and 50% without plaques) and ten-fold cross-validation were utilized to evaluate the model performance. Four machine learning methods (extreme gradient boosting (XGBoost), gradient boosting decision tree, random forest, and support vector machine) were evaluated. Subsequently, the interpretability of the XGBoost model, which provided the best performance, was analyzed from three aspects: feature importance, feature effect on prediction model, and feature effect on prediction decision for a specific subject.

RESULTS

The XGBoost algorithm provided the best performance (sensitivity: 0.8678, specificity: 0.8592, accuracy: 0.8632, F1 score: 0.8621, area under the curve: 0.8635) in carotid plaque prediction and also had excellent performance under missing data circumstances. Further, interpretability analysis showed that the decisions of the XGBoost model were highly congruent with clinical knowledge.

CONCLUSION

The model results are superior to those of state-of-the-art methods. Thus, it is a promising carotid plaque prediction tool that could be used in the primary prevention of stroke.

M., Arvind K. S.. Predictive Assessment of Fetus Features Using Scanned Image Segmentation Techniques and Deep Learning Strategy

Fetus weight at various stages of pregnancy is a critical component in determining the health of the baby. Abnormalities arising early in the pregnancy may be prevented by preventive measures. A variety of techniques suggested to predict foetus weight. Computer vision is a capability that can estimate the weight of a baby based on ultra-sonograms taken at various stages of pregnancy. Using the scanned data, one may train an advanced convolutional neural network that helps in accurately forecasting the fetus's size, weight, and overall health. The research utilizes computer vision techniques with image clustering methods for preprocessing, to predict the foetus's health, training datasets defective foetus datasets and healthy foetus datasets. Developing an integrated computer vision and a deep neural network is the hour which decrease the cost of operations and manual processes This study estimate the fetus's weight with optimal accuracy range at varying gestation age.

Data Enhancement and Deep Learning for Bone Age Assessment using The Standards of Skeletal Maturity of Hand and Wrist for Chinese

Conventional methods for artificial age determination of skeletal bones have several problems, such as strong subjectivity, large random errors, complex evaluation processes, and long evaluation cycles. In this study, an automated age determination of skeletal bones was performed based on Deep Learning. Two methods were used to evaluate bone age, one based on examining all bones in the palm and another based on the deep convolutional neural network (CNN) method. Both methods were evaluated using the same test dataset. Moreover, we can extend the dataset and increase the generalisation ability of the network by data expansion. Consequently, a more accurate bone age can be obtained. This method can reduce the average error of the final bone age evaluation and lower the upper limit of the absolute value of the error of the single bone age. The experiments show the effectiveness of the proposed method, which can provide doctors and users with more stable, efficient and convenient diagnosis support and decision support.

Comparison and development of advanced machine learning tools to predict nonalcoholic fatty liver disease: An extended study

BACKGROUND

Nonalcoholic fatty liver disease (NAFLD) is a public health challenge and significant cause of morbidity and mortality worldwide. Early identification is crucial for disease intervention. We recently proposed a nomogram-based NAFLD prediction model from a large population cohort. We aimed to explore machine learning tools in predicting NAFLD.

METHODS

A retrospective cross-sectional study was performed on 15 315 Chinese subjects (10 373 training and 4942 testing sets). Selected clinical and biochemical factors were evaluated by different types of machine learning algorithms to develop and validate seven predictive models. Nine evaluation indicators including area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve (AUPRC), accuracy, positive predictive value, sensitivity, F1 score, Matthews correlation coefficient (MCC), specificity and negative prognostic value were applied to compare the performance among the models. The selected clinical and biochemical factors were ranked according to the importance in prediction ability.

RESULTS

Totally 4018/10 373 (38.74%) and 1860/4942 (37.64%) subjects had ultrasound-proven NAFLD in the training and testing sets, respectively. Seven machine learning based models were developed and demonstrated good performance in predicting NAFLD. Among these models, the XGBoost model revealed the highest AUROC (0.873), AUPRC (0.810), accuracy (0.795), positive predictive value (0.806), F1 score (0.695), MCC (0.557), specificity (0.909), demonstrating the best prediction ability among the built models. Body mass index was the most valuable indicator to predict NAFLD according to the feature ranking scores.

CONCLUSIONS

The XGBoost model has the best overall prediction ability for diagnosing NAFLD. The novel machine learning tools provide considerable beneficial potential in NAFLD screening.

Prediction of Preterm Using Time Series Technology Based Machine Learning: Retrospective Cohort Study (Preprint)

Background

Objective

Methods

Results

Conclusions

Intelligent prediction of RBC demand in trauma patients using decision tree methods

BACKGROUND

The vital signs of trauma patients are complex and changeable, and the prediction of blood transfusion demand mainly depends on doctors' experience and trauma scoring system; therefore, it cannot be accurately predicted. In this study, a machine learning decision tree algorithm [classification and regression tree (CRT) and eXtreme gradient boosting (XGBoost)] was proposed for the demand prediction of traumatic blood transfusion to provide technical support for doctors.

METHODS

A total of 1371 trauma patients who were diverted to the Emergency Department of the First Medical Center of Chinese PLA General Hospital from January 2014 to January 2018 were collected from an emergency trauma database. The vital signs, laboratory examination parameters and blood transfusion volume were used as variables, and the non-invasive parameters and all (non-invasive + invasive) parameters were used to construct an intelligent prediction model for red blood cell (RBC) demand by logistic regression (LR), CRT and XGBoost. The prediction accuracy of the model was compared with the area under the curve (AUC).

RESULTS

For non-invasive parameters, the LR method was the best, with an AUC of 0.72 [95% confidence interval (CI) 0.657-0.775], which was higher than the CRT (AUC 0.69, 95% CI 0.633-0.751) and the XGBoost (AUC 0.71, 95% CI 0.654-0.756, P < 0.05). The trauma location and shock index are important prediction parameters. For all the prediction parameters, XGBoost was the best, with an AUC of 0.94 (95% CI 0.893-0.981), which was higher than the LR (AUC 0.80, 95% CI 0.744-0.850) and the CRT (AUC 0.82, 95% CI 0.779-0.853, P < 0.05). Haematocrit (Hct) is an important prediction parameter.

CONCLUSIONS

The classification performance of the intelligent prediction model of red blood cell transfusion in trauma patients constructed by the decision tree algorithm is not inferior to that of the traditional LR method. It can be used as a technical support to assist doctors to make rapid and accurate blood transfusion decisions in emergency rescue environment, so as to improve the success rate of patient treatment.

Prediction of fetal weight based on back propagation neural network optimized by genetic algorithm

Fetal weight is an important index to judge fetal development and ensure the safety of pregnant women. However, fetal weight cannot be directly measured. This study proposed a prediction model of fetal weight based on genetic algorithm to optimize back propagation (GA-BP) neural network. Using random number table method, 80 cases of pregnant women in our hospital from September 2018 to March 2019 were divided into control group and observation group, 40 cases in each group. The doctors in the control group predicted the fetal weight subjectively according to routine ultrasound and physical examination. In the observation group, the continuous weight change model of pregnant women was established by using the regression model and the historical physical examination data obtained by feature normalization pretreatment, and then the genetic algorithm (GA) was used to optimize the initial weights and thresholds of back propagation (BP) neural network to establish the fetal weight prediction model. The coincidence rate of fetal weight was compared between the two groups after birth. Results: The prediction error of GA-BPNN was controlled within 6%. And the accuracy of GA-BPNN was 76.3%, which were 14.5% higher than that of traditional methods. According to the error curve, GA-BP is more effective in predicting the actual fetal weight. Conclusion: The GA-BPNN model can accurately and quickly predict fetal weight.

Predicting fetal weight by three-dimensional limb volume ultrasound (AVol/TVol) and abdominal circumference

BACKGROUND

Fetal weight is an important parameter to ensure maternal and child safety. The purpose of this study was to use three-dimensional (3D) limb volume ultrasound combined with fetal abdominal circumference (AC) measurement to establish a model to predict fetal weight and evaluate its efficiency.

METHODS

A total of 211 participants with single pregnancy (28-42 weeks) were selected between September 2017 and December 2018 in the Beijing Obstetrics and Gynecology Hospital of Capital Medical University. The upper arm (AVol)/thigh volume (TVol) of fetuses was measured by the 3D limb volume technique. Fetal AC was measured by two-dimensional ultrasound. Nine cases were excluded due to incomplete information or the interval between examination and delivery >7 days. The enrolled 202 participants were divided into a model group (134 cases, 70%) and a verification group (68 cases, 30%) by mechanical sampling method. The linear relationship between limb volume and fetal weight was evaluated using Pearson Chi-squared test. The prediction model formula was established by multivariate regression with data from the model group. Accuracy of the model formula was evaluated with verification group data and compared with traditional formulas (Hadlock, Lee2009, and INTERGROWTH-21st) by paired t-test and residual analysis. Receiver operating characteristic curves were generated to predict macrosomia.

RESULTS

AC, AVol, and TVol were linearly related to fetal weight. Pearson correlation coefficient was 0.866, 0.862, and 0.910, respectively. The prediction model based on AVol/TVol and AC was established as follows: Y = -481.965 + 12.194TVol + 15.358AVol + 67.998AC, R2adj = 0.868. The scatter plot showed that when birth weight fluctuated by 5% (i.e., 95% to 105%), the difference between the predicted fetal weight by the model and the actual weight was small. A paired t-test showed that there was no significant difference between the predicted fetal weight and the actual birth weight (t = -1.015, P = 0.314). Moreover, the residual analysis showed that the model formula's prediction efficiency was better than the traditional formulas with a mean residual of 35,360.170. The combined model of AVol/TVol and AC was superior to the Lee2009 and INTERGROWTH-21st formulas in the diagnosis of macrosomia. Its predictive sensitivity and specificity were 87.5% and 91.7%, respectively.

CONCLUSION

Fetal weight prediction model established by semi-automatic 3D limb volume combined with AC is of high accuracy, sensitivity, and specificity. The prediction model formula shows higher predictive efficiency, especially for the diagnosis of macrosomia.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT03002246; https://clinicaltrials.gov/ct2/show/NCT03002246?recrs=e&cond=fetal&draw=8&rank=67.

Lung Nodule Detection based on Faster R-CNN Framework

BACKGROUND

Lung cancer is a worldwide high-risk disease, and lung nodules are the main manifestation of early lung cancer. Automatic detection of lung nodules reduces the workload of radiologists, the rate of misdiagnosis and missed diagnosis. For this purpose, we propose a Faster R-CNN algorithm for the detection of these lung nodules.

METHOD

Faster R-CNN algorithm can detect lung nodules, and the training set is used to prove the feasibility of this technique. In theory, parameter optimization can improve network structure, as well as detection accuracy.

RESULT

Through experiments, the best parameters are that the basic learning rate is 0.001, step size is 70,000, attenuation coefficient is 0.1, the value of Dropout is 0.5, and the value of Batch Size is 64. Compared with other networks for detecting lung nodules, the optimized and improved algorithm proposed in this paper generally improves detection accuracy by more than 20% when compared with the other traditional algorithms.

CONCLUSION

Our experimental results have proved that the method of detecting lung nodules based on Faster R-CNN algorithm has good accuracy and therefore, presents potential clinical value in lung disease diagnosis. This method can further assist radiologists, and also for researchers in the design and development of the detection system for lung nodules.

Convolutional neural network for automatically segmenting magnetic resonance images of the shoulder joint

BACKGROUND

Magnetic resonance imaging (MRI) has been known to replace computed tomography (CT) for bone and skeletal joint examination. The accurate automatic segmentation of bone structure in shoulder MRI is important for the measurement and diagnosis of bone injury and disease. Existing bone segmentation algorithms cannot achieve automatic segmentation without any prior knowledge, and their versatility and accuracy are relatively low. Therefore, an automatic segmentation combining pulse coupled neural network (PCNN) and full convolutional neural networks (FCN) is proposed.

METHODOLOGY

By constructing the block-based AlexNet segmentation model and U-Net-based bone segmentation module, we implemented the humeral segmentation model, articular bone segmentation model, humeral head and articular bone segmentation model synthesis model. We use this four kinds of segmentation models to obtain candidate bone regions, and accurately detect the positions of humerus and articular bone by voting. Finally, we perform an AlexNet segmentation model in the detected bone area in one step to segment accuracy at the pixel level.

RESULTS

The experimental data came from 8 groups of patients in Shengjing Hospital affiliated to China Medical University. The scanning volume of each group is approximately 100 images. Five fold cross-validations and for training were recorded, and five sets of data were carefully separated. After using our technique in the three groups of patients tested, the positive predictive value of dice coefficient (PPV) and the average accuracy of sensitivity were very significant, which reached 0.96±0.02, 0.97±0.02 and 0.94±0.03, respectively.

CONCLUSION

The method used in the experiment in this paper is based on a small amount of patient sample data. The deep learning required for the experiment needs to be performed through 2D medical images. The shoulder segmentation data obtained in this way can be very accurate.

Fetal birthweight prediction with measured data by a temporal machine learning method

Background

Birthweight is an important indicator during the fetal development process to protect the maternal and infant safety. However, birthweight is difficult to be directly measured, and is usually roughly estimated by the empirical formulas according to the experience of the doctors in clinical practice.

Methods

This study attempts to combine multiple electronic medical records with the B-ultrasonic examination of pregnant women to construct a hybrid birth weight predicting classifier based on long short-term memory (LSTM) networks. The clinical data were collected from 5,759 Chinese pregnant women who have given birth, with more than 57,000 obstetric electronic medical records. We evaluated the prediction by the mean relative error (MRE) and the accuracy rate of different machine learning classifiers at different predicting periods for first delivery and multiple deliveries. Additionally, we evaluated the classification accuracies of different classifiers respectively for the Small-for-Gestational-age (SGA), Large-for-Gestational-Age (LGA) and Appropriate-for-Gestational-Age (AGA) groups.

Results

The results show that the accuracy rate of the prediction model using Convolutional Neuron Networks (CNN), Random Forest (RF), Linear-Regression, Support Vector Regression (SVR), Back Propagation Neural Network(BPNN), and the proposed hybrid-LSTM at the 40th pregnancy week for first delivery were 0.498, 0.662, 0.670, 0.680, 0.705 and 0.793, respectively. Among the groups of less than 39th pregnancy week, the 39th pregnancy week and more than 40th week, the hybrid-LSTM model obtained the best accuracy and almost the least MRE compared with those of machine learning models. Not surprisingly, all the machine learning models performed better than the empirical formula. In the SGA, LGA and AGA group experiments, the average accuracy by the empirical formula, logistic regression (LR), BPNN, CNN, RF and Hybrid-LSTM were 0.780, 0.855, 0.890, 0.906, 0.916 and 0.933, respectively.

Conclusions

The results of this study are helpful for the birthweight prediction and development of guidelines for clinical delivery treatments. It is also useful for the implementation of a decision support system using the temporal machine learning prediction model, as it can assist the clinicians to make correct decisions during the obstetric examinations and remind pregnant women to manage their weight.

Towards deep phenotyping pregnancy: a systematic review on artificial intelligence and machine learning methods to improve pregnancy outcomes

Objective

Development of novel informatics methods focused on improving pregnancy outcomes remains an active area of research. The purpose of this study is to systematically review the ways that artificial intelligence (AI) and machine learning (ML), including deep learning (DL), methodologies can inform patient care during pregnancy and improve outcomes.

Materials and methods

We searched English articles on EMBASE, PubMed and SCOPUS. Search terms included ML, AI, pregnancy and informatics. We included research articles and book chapters, excluding conference papers, editorials and notes.

Results

We identified 127 distinct studies from our queries that were relevant to our topic and included in the review. We found that supervised learning methods were more popular (n = 69) than unsupervised methods (n = 9). Popular methods included support vector machines (n = 30), artificial neural networks (n = 22), regression analysis (n = 17) and random forests (n = 16). Methods such as DL are beginning to gain traction (n = 13). Common areas within the pregnancy domain where AI and ML methods were used the most include prenatal care (e.g. fetal anomalies, placental functioning) (n = 73); perinatal care, birth and delivery (n = 20); and preterm birth (n = 13). Efforts to translate AI into clinical care include clinical decision support systems (n = 24) and mobile health applications (n = 9).

Conclusions

Overall, we found that ML and AI methods are being employed to optimize pregnancy outcomes, including modern DL methods (n = 13). Future research should focus on less-studied pregnancy domain areas, including postnatal and postpartum care (n = 2). Also, more work on clinical adoption of AI methods and the ethical implications of such adoption is needed.

Cardiac magnetic resonance image segmentation based on convolutional neural network

OBJECTIVE

In cardiac medical imaging, the extraction and segmentation of the part of interest is the key to the diagnosis of heart disease. Due to irregular diastole and contraction, magnetic resonance imaging (MRI) images have poorly defined boundaries, and traditional segmentation algorithms have poor performance. In this paper, a cardiac MRI segmentation technique using convolutional neural network and image saliency is suggested.

METHODS

The convolutional neural network is used for detecting target area, filter out the ribs, muscles and the other parts of the anatomy where the contrast is not clearly defined. It can also be used to extract the region of interest (ROI), and compute the contrast of the ROI in order to improve clarity of the heart tissue within the ROI. The cardiac image diagnosis is performed using the obtained saliency image and compared with the segmentation result of the region growth algorithm. Finally, the images of 85 patients were used to train and test the algorithm model. Here, 46 patients were randomly selected for training, and the remaining 39 were harnessed for further tests.

RESULTS

Segmentation accuracy of our algorithm model in ventricles, septum and the apex of the heart segment reaches 93.14%, 92.58% and 96.21% respectively, which are better than the segmentation method based on the regional growth technique.

CONCLUSIONS

The segmentation method using convolutional neural network and image saliency can meet the needs of automatic heart segmentation tasks based on cardiac MRI image sequences. The segmented image is able to assist the doctor to observe the patient's heart health more effectively. As such, our proposed technique has strong potential in clinical applications.

Skeletal bone age prediction based on a deep residual network with spatial transformer

OBJECTIVE

Bone age prediction can be performed by medical experts manually assessment of X-ray images of the hand bone. In practice, the workload is huge, resource consumption is large, measurement takes a long time, and it is easily influenced by human factors. As such, manual estimation of bone age takes a long time and the results fluctuate greatly depending on the proficiency of the radiologist.

METHODS

The left-hand X-ray image data was identified and pre-processed. X-ray image analysis method using on deep neural network was used to automatically extract the key features of the left-hand joint bone age, and evaluation performance of the model was implemented.

RESULTS

In this paper, the deep learning method can be used to obtain the X-ray bone image features, and the convolutional neural network is used to automatically assess the age of bone. The feature region extraction method based on deep learning can extract feature information with superior performance compared to the traditional image analysis technique. Based on the residual network (ResNet) model in the deep learning algorithm, the average absolute error of the age of bones detected by the bone age assessment model is 0.455 better than traditional methods and only end-to-end deep learning methods. When the learning rate is greater than 0.0005, the MAE of Inception Resnet v2 model is higher than most models. Accuracy of bone age prediction is as high as 97.6%.

CONCLUSION

In comparison with the traditional machine learning feature extraction technique, the convolutional neural network based on feature extraction has better performance in the bone age regression model, and further improves the accuracy of image-based age of bone assessment.

A novel U-Net approach to segment the cardiac chamber in magnetic resonance images with ghost artifacts

OBJECTIVE

We propose a robust technique for segmenting magnetic resonance images of post-atrial septal occlusion intervention in the cardiac chamber.

METHODS

A variant of the U-Net architecture is used to perform atrial segmentation via a deep convolutional neural network, and we compare performance with the Kass snake model. It can be used to determine the surgical success of atrial septal occlusion (ASO) pre- and post- the implantation of the septal occluder, which is based on the volume restoration of the right atria (RA) and left atria (LA).

RESULTS

The method was evaluated on a test dataset containing 550 two-dimensional image slices, outperforming conventional active contouring regarding the Dice similarity coefficient, Jaccard index, and Hausdorff distance, and achieving segmentation in the presence of ghost artifacts that occlude the atrium outline. This problem has been unsolvable using traditional machine learning algorithm pertaining to active contouring via the Kass snake algorithm. Moreover, the proposed technique is closer to manual segmentation than the snakes active contour model in mean of atrial area (M-AA), mean of atrial maximum diameter (M-AMXD), mean atrial minimum diameter (M-AMID), and mean angle of the atrial long axis (M-AALA).

CONCLUSION

After segmentation, we compute the volume ratio of right to left atria, obtaining a smaller ratio that indicates better restoration. Hence, the proposed technique allows to evaluate the surgical success of atrial septal occlusion and may support diagnosis regarding the accurate evaluation of atrial septal defects before and after occlusion procedures.