Non-Invasive blood glucose monitoring using infrared (IR) light is considered to be a useful and reliable tool for measuring blood sugar levels during daily activities. IR-based glucose monitoring depends on the variant absorption levels of IR light waves by blood with high or low levels of glucose solution. This paper introduces a low-cost finger probe to measure glucose based on Arduino and embedding a Clarke error grid with fuzzy logic. An electronic blood glucose meter was designed and implemented in a non-invasive and painless manner based on an infrared sensor. The electrical signal expressing the level of glucose in the blood with a mathematical equation was used to calibrate and map the physical and electrical values. The final numerical value was validated with the Clarke error grid by implementing fuzzy logic (FL). The designed device was tested on 30 subjects with 15 diabetes subjects. The results show the high significance of results at points where the FL was able to determine an error range of less than 10% of measured glucose within the same range of the reference measurements. The proposed method of using FL with a Clarke error grid gives more confident and precise outputs in cases of this kind of portable device.

Breast cancer (BC) is a prominent cause of female mortality on a global scale. Recently, there has been growing interest in utilizing blood and tissue-based biomarkers to detect and diagnose BC, as this method offers a non-invasive approach. To improve the classification and prediction of BC using large biomarker datasets, several machine-learning techniques have been proposed. In this paper, we present a multi-stage approach that consists of computing new features and then sorting them into an input image for the ResNet50 neural network. The method involves transforming the original values into normalized values based on their membership in the Gaussian distribution of healthy and BC samples of each feature. To test the effectiveness of our proposed approach, we employed the Coimbra and Wisconsin datasets. The results demonstrate efficient performance improvement, with an accuracy of 100% and 100% using the Coimbra and Wisconsin datasets, respectively. Furthermore, the comparison with existing literature validates the reliability and effectiveness of our methodology, where the normalized value can reduce the misclassified samples of ML techniques because of its generality.

Diabetes mellitus (DM) is a chronic metabolic condition characterized by high blood glucose levels owing to decreased insulin production or sensitivity. Current diagnostic approaches for gestational diabetes entail intrusive blood tests, which are painful and impractical for regular monitoring. Additionally, typical blood glucose monitoring systems are restricted in their measurement frequency and need finger pricks for blood samples. This research study focuses on the development of a non-invasive, real-time glucose monitoring method based on the detection of glucose in human tears and finger blood using mid-infrared (IR) spectroscopy. The proposed solution combines a fuzzy logic-based calibration mechanism with an IR sensor and Arduino controller. This calibration technique increases the accuracy of non-invasive glucose testing based on MID absorbance in fingertips and human tears. The data demonstrate that our device has high accuracy and reliability, with an error rate of less than 3%, according to the EGA. Out of 360 measurements, 97.5% fell into zone A, 2.2% into zone B, and 0.3% into zone C of the Clarke Error Grid. This suggests that our device can give clinically precise and acceptable estimates of blood glucose levels without inflicting any harm or discomfort on the user.

Infrared neuromodulation (INM) is a promising neuromodulation tool that utilizes pulsed or continuous-wave near-infrared (NIR) laser light to produce an elevation of the background temperature of the neural tissue. The INM-based cortical heating has been proven as an effective modality to induce changes in neuronal activities. In this paper, we investigate the effect of INM-based cortical heating on the characteristics of interictal epileptiform discharges (IEDs) induced by penicillin in anesthetized rats. Cortical heating was conducted using a NIR laser light guided through a needle-like silicon-based waveguide probe. We detected penicillin-induced cortical IEDs from preprocessed micro-electrocorticography ( $$\mu$$μ ECoG) recordings, then we assessed changes in various temporal and spectral features of IEDs due to INM. Our findings show that the fast cortical heating phase obtained with continuous-wave NIR light is highly associated with a reduction of IED amplitudes, small but significant changes in the negative amplitude of IEDs compared with the baseline, and a proportional increase in the power of frequency bands related to delta/theta (2–8 Hz) and gamma (28–80 Hz) oscillations. Furthermore, a low rate of cortical heating with pulsed NIR illumination has a more inhibitory impact on the sharp negative polarity of IEDs. Our findings do not indicate a clear reduction in the frequency of IEDs in anesthetized rodents. In contrast, 2–4 min of continuous laser illumination leads to a notable increase in IED frequency. This effect of INM could potentially restrict its use in therapeutic applications related to epilepsy. However, the thermal effect of INM on cortical neurons induces changes in other characteristics of IEDs, which could prove beneficial for future applications.

Electrochemical impedance spectroscopy (EIS) is the golden tool for many emerging biomedical applications that describes the behavior, stability, and long-term durability of physical interfaces in a specific range of frequency. Impedance measurements of any biointerface during in vivo and clinical applications could be used for assessing long-term biopotential measurements and diagnostic purposes. In this paper, a novel approach to predicting impedance behavior is presented and consists of a dimensional reduction procedure by converting EIS data over many days of an experiment into a one-dimensional sequence of values using a novel formula called day factor (DF) and then using a long short-term memory (LSTM) network to predict the future behavior of the DF. Three neural interfaces of different material compositions with long-term in vitro aging tests were used to validate the proposed approach. The results showed good accuracy in predicting the quantitative change in the impedance behavior (i.e., higher than 75%), in addition to good prediction of the similarity between the actual and the predicted DF signals, which expresses the impedance fluctuations among soaking days. The DF approach showed a lower computational time and algorithmic complexity compared with principal component analysis (PCA) and provided the ability to involve or emphasize several important frequencies or impedance range in a more flexible way.

Sensing insole systems are a promising technology for various applications in healthcare and sports. They can provide valuable information about the foot pressure distribution and gait patterns of different individuals. However, designing and implementing such systems poses several challenges, such as sensor selection, calibration, data processing, and interpretation. This paper proposes a sensing insole system that uses force-sensitive resistors (FSRs) to measure the pressure exerted by the foot on different regions of the insole. This system classifies four types of foot deformities: normal, flat, over-pronation, and excessive supination. The classification stage uses the differential values of pressure points as input for a feedforward neural network (FNN) model. Data acquisition involved 60 subjects diagnosed with the studied cases. The implementation of FNN achieved an accuracy of 96.6% using 50% of the dataset as training data and 92.8% using only 30% training data. The comparison with related work shows good impact of using the differential values of pressure points as input for neural networks compared with raw data.