Knee osteoarthritis (KOA) is a leading cause of disability, particularly affecting older adults due to the deterioration of articular cartilage within the knee joint. This condition is characterized by pain, stiffness, and impaired movement, posing a significant challenge in medical diagnostics and treatment planning, especially due to the current inability for early and accurate detection or monitoring of disease progression. This research introduces a multifaceted approach employing feature extraction and machine learning (ML) to improve the accuracy of diagnosing and classifying KOA stages from radiographic images. Utilizing a dataset of 3154 knee X-ray images, this study implemented feature extraction methods such as Histogram of Oriented Gradients (HOG) with Linear Discriminant Analysis (LDA) and Min–Max scaling to prepare the data for classification. The study evaluates six ML classifiers—K Nearest Neighbors classifier, Support Vector Machine (SVM), Gaussian Naive Bayes, Decision Tree, Random Forest, and XGBoost—optimized via GridSearchCV for hyperparameter tuning within a 10-fold Stratified K-Fold cross-validation framework. An ensemble model has also been made for the already high-accuracy models to explore the possibility of enhancing the accuracy and reducing the risk of overfitting. The XGBoost classifier and the ensemble model emerged as the most efficient for multiclass classification, with an accuracy of 98.90%, distinguishing between healthy and unhealthy knees. These results underscore the potential of integrating advanced ML methodologies for the nuanced and accurate diagnosis and classification of KOA, offering new avenues for clinical application and future research in medical imaging diagnostics.

Regular monitoring of blood α-fetoprotein (AFP) and/or carcino-embryonic antigen (CEA) levels is important for the routine screening of liver cancer. However, AFP and CEA have a much lower specificity than des-γ-carboxyprothrombin (DCP) to detect liver cancer. Therefore, the study reported here was designed, to develop a screen-printed DCP immunosensor incorporating zinc oxide nanoparticles, for accurate determination of DCP. The designed immunosensor shows low detection limits for the detection of DCP: 0.440 ng/mL (based on impedance measurement), 0.081 ng/mL (based on real part of impedance measurement) and 0.078 ng/mL (based on imaginary part of impedance measurement), within the range of 3.125 ng/mL to 2000 ng/mL. In addition, there was little interference to DCP determination by molecules such as Na⁺, K⁺, Ca²⁺, Cl⁻, glucose, urea, and uric acid. It is therefore concluded that the DCP immunosensor developed and reported here is simple, inexpensive and effective, and shows promise in the rapid screening of early-stage liver cancer at home with a point-of-care approach.

Artificial intelligence (AI) radar technology offers several advantages over other technologies, including low cost, privacy assurance, high accuracy, and environmental resilience. One challenge faced by AI radar technology is the high cost of equipment and the lack of radar datasets for deep-learning model training. Moreover, conventional radar signal processing methods have the obstacles of poor resolution or complex computation. Therefore, this paper discusses an innovative approach in the integration of radar technology and machine learning for effective surveillance systems that can surpass the aforementioned limitations. This approach is detailed into three steps: signal acquisition, signal processing, and feature-based classification. A hardware prototype of the signal acquisition circuitry was designed for a Continuous Wave (CW) K-24 GHz frequency band radar sensor. The collected radar motion data was categorized into non-human motion, human walking, and human walking without arm swing. Three signal processing techniques, namely short-time Fourier transform (STFT), mel spectrogram, and mel frequency cepstral coefficients (MFCCs), were employed. The latter two are typically used for audio processing, but in this study, they were proposed to obtain micro-Doppler spectrograms for all motion data. The obtained micro-Doppler spectrograms were then fed to a simplified 2D convolutional neural networks (CNNs) architecture for feature extraction and classification. Additionally, artificial neural networks (ANNs) and 1D CNN models were implemented for comparative analysis on various aspects. The experimental results demonstrated that the 2D CNN model trained on the MFCC feature outperformed the other two methods. The accuracy rate of the object classification models trained on micro-Doppler features was 97.93%, indicating the effectiveness of the proposed approach.

Day-old male chick culling is one of the world’s most inhumane problems in the poultry industry. Every year, seven billion male chicks are slaughtered in laying-hen hatcheries due to their higher feed exchange rate, lower management than female chicks, and higher production costs. This study describes a novel non-invasive method for determining the gender of chicken eggs. During the incubation period of fourteen days, four electrodes were attached to each egg for data collection. On the last day of incubation, a standard polymerase chain reaction (PCR)-based chicken gender determination protocol was applied to the eggs to obtain the gender information. A relationship was built between the collected data and the egg’s gender, and it was discovered to have a reliable connection, indicating that the chicken egg gender can be determined by measuring the impedance data of the eggs on day 9 of incubation with the four electrodes set and using the self-normalization technique. This is a groundbreaking discovery, demonstrating that impedance spectroscopy can be used to sex chicken eggs before they hatch, relieving the poultry industry of such an ethical burden.

The label-free biosensor has emerged as an effective tool for the purpose of early detection of causative pathogens such as Escherichia coli as a preventive measure. In this study, a biorecognition-element-free interdigitated microelectrode (IDμE) sensor is designed and developed with this in mind, with good reliability and affordability. Results show that the designed sensor can identify E. coli with good selectivity using an impedance and capacitance of 7.69 MHz. At its optimum impedance of 1.3 kHz, the IDμE sensor can reliably quantify E. coli in a range of measurement (10^3.2~10⁶ cfu/mL), linearity (R² = 0.97), sensitivity (18.15 kΩ/log (cfu/mL)), and limit of detection (10^3.2 cfu/mL). In summary, the IDμE sensor developed possesses high potential for industrial and clinical applications.

Pancreatic cancer (PC) is a global health problem that features a very high mortality rate. The UL16 binding protein 2 (ULBP2) is a new biomarker for PC detection. This study develops a simple, reliable, and inexpensive immunosensor for the detection of the ULBP2 antigen while also investigating the effects of an array configuration of connected sensors and zinc oxide (ZnO) nanoparticles on the immunosensor’s sensitivity. The ULBP2 antibody was immobilized onto the screen-printed carbon electrode (SPCE) surfaces of three different sensors: a simple SPCE (ULBP2-SPCE); an SPCE array, which is a series of identical SPCE connected to each other at different arrangements of rows and columns (ULBP2-SPCE-1x2 and ULBP2-SPCE-1x3); and an SPCE combined with ZnO nanoparticles (ULBP2-ZnO/SPCE). Impedance spectrum measurements for the immunosensors to ULBP2 antigen were conducted and compared. According to the result, the array configurations (ULBP2-SPCE-1x2 and ULBP2-SPCE-1x3) show an improvement of sensitivity compared to the ULBP2-SPCE alone, but the improvement is not as significant as that of the ULBP2-ZnO/SPCE configuration (ULBP2-ZnO/SPCE > ULBP2-SPCE: 18 times larger). The ULBP2-ZnO/SPCE immunosensor has a low limit of detection (1 pg/mL) and a high sensitivity (332.2 Ω/Log(pg/mL)), excellent linearity (R² = 0.98), good repeatability (coefficients of variation = 5.03%), and is stable in long-term storage (retaining 95% activity after 28 days storage). In an array configuration, the immunosensor has an increased signal-to-noise ratio (ULBP2-SPCE-1x3 > ULBP2-SPCE: 1.5-fold) and sensitivity (ULBP2-SPCE-1x3 > ULBP2-SPCE: 2.6-fold). In conclusion, either the modification with ZnO nanoparticles onto the sensor or the use of an array configuration of sensors can enhance the immunosensor’s sensitivity. In this study, the best immunosensor for detecting ULBP2 antigens is the ULBP2-ZnO/SPCE immunosensor.