Artificial intelligence is serving as an impetus in digital health, clinical support, and health informatics for an informed patient’s outcome. Previous studies only consider classification accuracies of cardiotocographic (CTG) datasets and disregard computational time, which is a relevant parameter in a clinical environment. This paper proposes a modified deep neural algorithm to classify untapped pathological and suspicious CTG recordings with the desired time complexity. In our newly developed classification algorithm, AlexNet architecture is merged with support vector machines (SVMs) at the fully connected layers to reduce time complexity. We used an open-source UCI (Machine Learning Repository) dataset of cardiotocographic (CTG) recordings. We divided 2126 CTG recordings into 3 classes (Normal, Pathological, and Suspected), including 23 attributes that were dynamically programmed and fed to our algorithm. We employed a deep transfer learning (TL) mechanism to transfer prelearned features to our model. To reduce time complexity, we implemented a strategy wherein layers in the convolutional base were partially trained to leave others in the frozen states. We used an ADAM optimizer for the optimization of hyperparameters. The presented algorithm also outperforms the leading architectures (RCNNs, ResNet, DenseNet, and GoogleNet) with respect to real-time accuracies, sensitivities, and specificities of 99.72%, 96.67%, and 99.6%, respectively, making it a viable candidate for clinical settings after real-time validation.

/pdf/accessing-artificial-intelligence-for-fetus-health-status-18m6x0li.pdf

Accessing Artificial Intelligence for Fetus Health Status Using Hybrid Deep Learning Algorithm (AlexNet-SVM) on Cardiotocographic Data

Ultrasound imaging is a non-invasive approach for diagnosing internal organs that is both safe and effective. Fetal biometric measurement is a routine examination performed during pregnancy to evaluate fetal growth and estimate gestational age and weight using ultrasound images. The prenatal diagnosis of fetus helps to identify whether the growth of fetus is normal or abnormal. There are many biometric measurements like femur length, head circumference (HC), biparietal diameter(BPD), abdominal circumference etc. In this work, Head circumference (HC) measurement is performed. Head circumference (HC) is measured as a critical biometric to assess fetus growth throughout the pregnancy. Ultrasound scans are used to perform this measurement. The dataset used here is HC18 grand challenge. U-Net and Attention U-Net are used for segmenting the fetal head portion. The trained model undergoes Morphological processing and edge detection, then the head portion are fitted to the ellipse and the circumference is measured. The metrics used are Mean square error, E-Score, Accuracy and IoU score to evaluate the result.

Fetal Head Circumference Measurement from Ultrasound Images Using Attention U-Net

Demystifying evidential Dempster Shafer-based CNN architecture for fetal plane detection from 2D ultrasound images leveraging fuzzy-contrast enhancement and explainable AI.

Monitoring fetal growth is an essential activity throughout gestation period. It is necessary for the well-being of the mother and the child so that any abnormalities can be diagnosed at the earlier stage itself and preventive measures can be taken. MRI Image analysis is necessary to identify the anomalies that have a higher probability of occurrence during the first and second trimesters. Fetal brain MRI images are obtained and analysed for any disorder in the neurodevelopment of the embryo. Classification of the fetal brain image into normal and abnormal cases is done using deep learning algorithms like densenet121 and squeezenet1_0. Densenet produced an accuracy of 98.7% when compared to squeezenet whose accuracy is 96.3%. The performance of densenet is high in case of various metrics like precision, recall, sensitivity and F1 Score.

Performance Analysis of Squeezenet and Densenet on Fetal Brain MRI Dataset

Road damage occurs when the function and structure of road are unable to service the traffic above it optimally. In general, the damage is caused by flaws in planning and implementation, uneven maintenance, poor drainage, and poor road user behaviour. It has a negative impact on driving comfort, road safety, and vehicle condition, and it may cause a number of accidents. To address this issue, this study presents a Region-based Convolutional Neural Network (R-CNN) for locating the dangerous path. This type of neural network can find essential information in both time series and picture data is the RCNN. As a result, it is extremely useful for image-associate tasks including image identification, object categorization, and design recognition. A RCNN uses linear algebra methods such as matrix multiplication to discover patterns inside an image. Find the photographs first and pre-process them, then extract the features and choose them from the feature set of previously damaged images. Finally, categorise the captured photos to obtain the optimum result. When compared to other current approaches, the suggested method is more accurate.

Parvathavarthini S

Papers

Performance Analysis of Squeezenet and Densenet on Fetal Brain MRI Dataset

Road Damage Detection using Deep Learning