Biomedical Engineering: Applications, Basis and Communications

The medical image can be divided into two major classes: anatomical image and functional image. Magnetic resonance image (MRI) and computed tomography (CT) belong to the anatomical image, which can show an outline of an organ and a tissue clearly. The functional image can show the image of the organ and the tissue metabolism situation. Positron emission tomography (PET) and single photon emission computed tomography (SPECT) are also part of the functional image. In this paper, we use 4-[${}^{18}$F]-ADAM to measure serotonin transporter changes and find the relation with major depressive disorder in CT/PET three-dimensional image. In clinical diagnosis, based on the video imaging methods can be divided into the anatomy and function of these two categories. Therefore, the combination of anatomical and functional imaging can provide more useful information. In the study, we use the CT image to get the reference position and combine it with the isotope metabolic situation from PET. We use the fused image to construct 3D image and count serotonin transporter with different times, and measure the variance about serotonin transporter changes in the organization of the metabolism and distribution. We use the variance to find the relation and compare the score using Hamilton Depression Rating Scale by experienced psychiatrists. After the test cases, we measure the serotonin transporter distribution of depressed patients can do a preliminary classification. We can get the gray-level value changes in major depressive disorder patients at different times. The advantage of this method is to offer more information to diagnose or analyze the difference between the brain organs, such as pons, midbrain, thalamus, corpus striatum, and prefrontal cortex.

Lung cancer is one of the dangerous diseases that cause huge cancer death worldwide. Early detection of lung cancer is the only possible way to improve a patient’s chance for survival. This study presents an innovative automated diagnosis classification method for Computed Tomography (CT) images of lungs. In this paper, the CT scan of lung images was analyzed with the multiscale convolution. The entire lung is segmented from the CT images and the parameters are calculated from the segmented image. The use of image processing techniques and identifying patterns in the detection of lung cancer from CT images reduces human errors in detecting tumors, and speeds up diagnosis time. Artificial Neural Network (ANN) has been widely used to detect lung cancer, and has significantly reduced the percentage of errors. Therefore, in this paper, Convolution Neural Network (CNN), which is the most effective method, is used for the detection of various types of cancers. This study presents a Multiscale Convolutional Neural Network (MCNN) approach for the classification of tumors. Based on the structure of MCNN, which presents CT picture to several deep convolutional neural networks with different size and resolutions, the classical handcrafted features extraction step is avoided. The proposed approach gives better classification rates than the classical state of the art methods allowing a safer Computer-Aided Diagnosis of pleural cancer. This study reaches a diagnosis accuracy of $93.7\pm 0.3$ using multiscale convolution technique, which reveals the efficiency of the proposed method.

Nuclear medicine could provide an accurate estimate of the biochemical composition, metabolism, and capability of the kidney. Using nuclear medicine imaging to scan the organ, we can find the time-activity curve, called a renogram, by the intensity of the image. The activity curve can be employed to describe organ capability. Traditional methods of observing the activity curve of the whole kidney can obscure some small details. This paper proposes an improved renogram based on blood vessel distribution to estimate the kidney capability automatically. Every patient must acquire 180 renal dynamic images from the renal scan. First, acquiring region of the kidney in the No. 68 kidney image and correcting the background of this sample image, the sample image will be applied to the rest of the images to find the region of the kidney. The kidney image is then divided into several sub-regions according to the paths of the blood vessels from No. 50 kidney image. For each sub-region, we find the time–intensity curve, called the sub-renogram. Then, we observe the activity curves of the sub-region and the whole kidney separately to assist the diagnosis. The results of the automatic analysis are similar to the traditional methods used on the whole kidney, which ignores any abnormal sub-region activity. The proposed method can improve the accuracy of the doctor’s diagnosis.

Since there is an emergency to find the soldiers/human beings who are alive at a distance in the situation of wars/natural disasters, heart rate (HR) measurement using radar is a novel approach. Antennas installed with doppler radar, which radiates and receives electromagnetic (EM) signals, are used for detecting heart rate vitality with the non-contact approach. In the previous literature, different antenna’s like metamaterial leaky-wave antenna, rectangular patch antenna with different polarizations as transmitter and receiver, and inverted F patch antenna were used to detect the HR. Each of these methods has its own merits and demerits. In this paper, a review has been done on the antenna design to detect the heart vibrations which infers the cardiac conduction rate in a non-contact approach along with its signal processing work for HR measurement. The design issues in the literature and their effects of the non-contact approach on the human body were discussed. Finally, the review suggested possible future enhancements for improving the performance of HR measurement using a non-contact approach. This paper concludes that the HR measurement can be done at 2.4GHz/24GHz (ISM) frequency band, which comes in the S-band. Also, near field and far-field observations have been reported to know the performance of the design.

Cradle is a household appliance that carries the baby and aids in the comfortable sleep of the infants. In the current scenario, almost 80% of women are working. They find it tough to manage both the household work and office job. Hence, there is a real need for the design and development of a low-cost automatic oscillating cradle that could monitor the real-time parameters of the infants. Unlike adults, infants cannot regulate their body temperature easily. Children are more prone to develop hypothermia and hyperthermia under extreme temperature conditions. An incubator could maintain appropriate conditions for the infant. This study is focused on developing a low-cost automated baby cradle with an incubator that analyzes baby cry and oscillates automatically. The developed system can also maintain suitable environmental conditions for the infant’s growth. If in case attention of a premature baby were to be sought on a regular basis, the developed system can monitor the temperature and heartbeat along with the cabin temperature and humidity. The developed system has an inbuilt alarm that rings when there is an abnormality in the infant’s heartbeat and body temperature. The alarm also indicates when the mattress is wet. A Bluetooth-based mobile application is also designed which could monitor and control the cradle. The proposed prototype model can be employed both in hospitals and at home.

Biomedical color (BMC) images are being used on a wide scale by physicians, where their diagnosis would be more accurate. Hence, it is recommended to develop new approaches that are able to represent and retrieve the BMC images efficiently. This work proposes two methods to represent BMC images: Quaternion Associated Laguerre. Moments (Q_ALMs), and Quaternion Chebyshev Moments (Q_CMs). Q_ALMs and Q_CMs are derived by extending the ALMs and CMs to the quaternion field. ALMs and CMs represent discrete orthogonal moments, and they are defined using the Associated Laguerre Polynomials (ALPs) and Chebychev Polynomials, respectively. Hospitals and medical institutes everywhere in the world create and store a large variety of datasets of BMC images during the routine clinical practices; hence, the mastery to retrieve the BMC images correctly is crucial for precise diagnoses and also for the researchers in medical sciences. So that in this study, we also introduced two image retrieval systems for BMC images based on the Q_CMs and Q_ALMs approaches. Our approaches extensively assessed with two standard benchmark datasets: LGG Segmentation dataset for brain magnetic resonance MR images and NEMA-CT for the computed tomography (CT) images. The performance of the proposed retrieval systems is assessed through three performance metrics: Average retrieval precision (ARP), average retrieval rate (ARR), and F_score. Results have shown the outperformance of Q_CMs over Q_ALMs in both the cases of representing and retrieval of BMC images.

In this paper, an artificial intelligent technique is proposed for skin disease detection and classification. The suggested method comprises four stages, including segmentation, extraction of textural features, and classification. The stretch-based enhanced algorithm has been adapted for image enhancement. Then the method of an active contour is used for segmentation to determine the skin lesion in tissue. Textural features are obtained from the segmented skin lesion. As several numbers of the features can affect the classification precision, ideal feature selection is made to exclude features that are less informative and unnecessary. The feature selection is adjusted with a regularized random forest. Finally, the classification algorithms by support vector machine and a back-propagation neural network (BPNN) are implemented. The dataset consists of 400 dermoscopic images in total divided into 200 benign and 200 malignant skin diseases extracted from the dermoscopic images PH² database. The result of detecting and classifying the dermoscopic images on these images yielded an accuracy of 99.7%, a sensitivity of 99.4%, and a specificity of 100% by BPNN. The experiential results confirmed that the BPNN classifier is best rather than an SVM classifier for skin disease images. This proposed model will be advanced to support the skin image processing techniques that provided a more accurate diagnosis and rapid treatment plan.

Pedestrians, motorcyclists, and occupants of vehicle involved in side impacts can suffer lateral impacts at the shoulder with sufficiently high forces to fracture the clavicle. While some previous studies have developed a force tolerance for the clavicle, this study examines the clavicle under compression forces, both under dynamic and quasi-static loading conditions and then the experimental fracture patterns are compared with those identified in actual road traffic crashes. The quasi-static and dynamic tests were conducted in 24 clavicles from 14 donors aged 14–56 (11 males and 3 females, mean age $44.4\pm 13.2$). Clavicles were tested under compression forces until complete fracture. Probability of fracture is around 50% for 2335N in the quasi-static case (low strain rate) and 3045N in the dynamic case (high strain rate). In addition, risk curves for fracture have been computed from the sample. The experiments showed large differences in the peak force both in quasi-static ($2416\pm 781$N) and dynamic ($3183\pm 981$N) tests, but the displacement at the peak force was very similar ($4.9\pm 1.1$mm in quasi-static and $5.0\pm 1.3$mm in dynamic tests). Thus, a stiffer behavior of the specimens was observed for the dynamic tests. The Probability Density Function for dynamic and quasi-static cases differ significantly ($p$-value $<0.04$). The fracture patterns were consistent with principles of structural mechanics, predicting that the fracture will occur at points near to the highest bending moment point, and with the fractures observed in real road crashes.