Medical images can be constructed in two-dimensional (2D) or three-dimensional (3D) view imaging to be applied in disease detection and diagnosis inside the body, such as cancer/tumor, heart, or lung-related diseases. These images may contain patients’ privacy information and clinical diagnosis records. Hence, these images need to ensure authorization demands among hospitals, medical service organizations, or physicians in a picture archiving and communication system. This study presents an intelligent symmetric cryptography with a chaotic map and quantum-based key generator (KG) for medical image encryption and decryption. Overall scheme processes include (1) random cipher code generation, (2) training gray relational analysis (GRA)-based encryptor and decryptor, and (3) decrypted image evaluation. The hybrid chaotic map and quantum-based KG are used to increase the chaotic complexity and unpredictable levels to produce cipher codes for changing pixel values (substitution method) in a 2D image by 256 key-space cipher codes. The first and second GRA models are used to train the cipher codes to achieve an encryptor and a decryptor, respectively. Through the methodology validation using a chest X-ray database, the structural similarity index measurement is employed to evaluate the decryption quality between the plain image and decrypted image. The encrypted images show a visual uncorrelation with the plain images, and experimental results indicate higher confidences against the passive eavesdropper.

Lung diseases can result in acute breathing problems and prevent the human body from acquiring enough oxygen. These diseases, such as pneumonia (P), pleural effusion (Ef), lung cancer, pneumothorax (Pt), pulmonary fibrosis (F), infiltration (In) and emphysema (E), adversely affect airways, alveoli, blood vessels, pleura and other parts of the respiratory system. The death rates of P and lung cancer are higher than those of other typical lung diseases. In visualization examination, chest radiography, such as anterior-posterior or lateral image viewing, is a straightforward approach used by clinicians/radiologists to diagnose and locate possible lung abnormalities rapidly. However, a chest X-ray image of patients may show multiple abnormalities associated with coexisting conditions, such as P, E, F, Pt, atelectasis, lung cancer or surgical interventions, which further complicate diagnosis. In addition, poor-quality X-ray images and manual inspection have limitations in digital image-automated classification. Hence, this study intends to propose a multilayer machine vision classifier to automatically identify the possible class of lung diseases within a bounding region of interest (ROI) on a chest X-ray image. For digital image texture analysis, a two-dimensional (2D) fractional-order convolution (FOC) operation with a fractional-order parameter, $v =0.3-0.5$ , is used to enhance the symptomatic feature and remove unwanted noises. Then, maximum pooling is performed to reduce the dimensions of feature patterns and accelerate complex computations. A multilayer machine vision classifier with radial Bayesian network and gray relational analysis is used to screen subjects with typical lung diseases. Anterior-posterior chest X-ray images from the NIH chest X-ray database (NIH Clinical Center) are enrolled. For digital chest X-ray images, with $K$ -fold cross-validation, the proposed multilayer machine vision classifier is applied to facilitate the diagnosis of typical lung diseases on specific bounding ROIs, as promising results with mean recall (%), mean precision (%), mean accuracy (%) and mean F1 score of 98.68%, 82.42%, 83.57% and 0.8981, respectively, for assessing the performance of proposed multilayer classifier for rapidly screening lung lesions on digital chest X-ray images.

Digital physiological signals in telecare medicine information systems have been widely applied in remote medical applications, such as telecare, tele-examination, and telediagnosis, via computer networking transmission or wireless communication. However, these medical records need to ensure authorization demands in the channel model for human body communication and remote medical servers and enhance the confidentiality, recoverability, and availability of transmission data. Hence, this study proposes a symmetric cryptography scheme with a chaotic map and a multilayer machine learning network (MMLN) to achieve physiological signal infosecurity. A chaotic pseudorandom number generator within specific control parameters can dynamically produce unordered sequence numbers to set the secret keys for a regular secret key update, thereby improving the security of private cipher codes. The chaotic map is quickly iterated to produce a pseudorandom key stream for real-time applications, and the private cipher codes are selected using the initial and specific control parameters at the data emitter and receiver ends. A general regression neural network is used to map the high-dimensional input-output pair of cipher codes for substitution and permutation processes. Its adaptive MMLN with an optimization algorithm can rapidly train the random cipher code protocol to achieve an encryptor and a decryptor for a regular encrypted communication. Using the Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) Arrhythmia Database, 100 electrocardiogram fragments are used to verify the proposed model, and the peak signal-to-noise ratio (PSNR) as a quantitative quality metric is used to evaluate the visual quality after encryption and decryption processes for further diagnosis applications. Experimental results show that the proposed scheme has a higher mean PSNR (35.26 ± 3.77 dB) and shorter mean executing time (0.16 ± 0.01 s) compared with traditional cryptography protocol schemes.

Cardiomegaly is an asymptomatic disease. Symptoms, such as palpitations, chest tightness, and shortness of breath, may be the early indications of cardiac hypertrophy, which can be divided into cardiac hypertrophy and ventricular enlargement. Their causes and treatment strategies are different. The early detection of cardiomegaly can help to make decisions for administering drugs and surgical treatments. In addition, with regard to problems in manual inspection, such as time consuming and the need for human interpretations and experiences, an assistive tool is required to automatically develop and identify normal heart or enlarged hearts. Therefore, this study proposes the combination of 2D (two dimensional) and 1D (one dimensional) convolutional neural network based classifier for rapid cardiomegaly screening in clinical applications based on chest X-ray (CXR) examinations in frontal posteroanterior view. The 2D and 1D convolutional processes and multilayer connected classification network are used to enhance the original CXR image and to remove unwanted noises to increase accuracy in feature extraction and pattern recognition tasks. The training dataset and testing dataset are collected from the National Institutes of Health CXR image database, which is used to train the classifier and validate the performance of the classifier in a K-fold cross-validation manner. Experimental results indicate the potential performance for rapid cardiomegaly screening with regard to recall (%), precision (%), accuracy (%), and F1 score.

Consensus criteria for tremor classification in Parkinson’s disease (PD) patients are clinically important for automatically evaluating the PD rating scale. Wearable sensing tools with direct contact measurements can obtain physiological signals to monitor tremor symptoms. Then, machine learning algorithms (MLAs) can train the frequency-based parameters and motion features to accurately measure PD-related tremors. Noncontact measurement with customized computer information devices can also digitalize the digitized handwritten patterns with the bespoke movements that a hand makes for identifying tremor classes. However, wearable sensors need a set of multiple electrodes to be placed on a patient’s body to acquire biosignals, and the setup does not allow continuous measurement and limits the patient’s motion range. The handwritten patterns of noncontact-based methods need frequency-domain and linearization transformations. In addition, feature extraction methods and MLAs are limited in complex computations and adaptive applications. Hence, in this work, a noncontact measurement with an array X-band microwave (10 GHz) Doppler-based linear quantizer is designed to continuously measure upper limb movements for tremor class scaling. To overcome the complex computations, time-domain parametric features, including zero crossing (ZC), Willison amplitude (WAMP), and waveform length (WL) indexes, are used to extract the physical changes in the oscillation frequencies, amplitudes, and directions of tremor signals for scaling upper limb tremor (ULT) levels. In the experiments involving 10 subjects, the proposed noncontact bioradar sensor could quantify asymmetrical and irregular oscillations with a positive correlation (mean R ² > 0.85) between the three indexes (ZC, WAMP, WL) and various oscillation frequencies. The linear relationship quantizer could predict the ULT levels from 0 Hz to 8 Hz for PD patients (typical tremor frequency: 4–6 Hz). It could also map the three indexes into colored visual representation for computerized visual analysis.