In general, epilepsy is considered as one of most prevalent neurological disorders and frequently appears as sudden seizures, resulting in injuries, accidents, sudden unexpected deaths, etc. It is reported that around 60 million people across the globe experience various seizures due to epilepsy. So, there is a demand for ambulatory seizure detection devices to prevent such accidents and to improve the quality of life of epilepsy patients. In this work, an intelligent Internet of Medical Things (IoMT)-based wearable device is designed and developed to monitor seizures in epilepsy patients. Due to the lack of an accelerometer dataset for epileptic seizures, the proposed device is developed, and a dataset simulating seizure-like activities has been generated. Further, the proposed device utilises an MPU6500-based Inertial Measurement Unit (IMU), which is integrated to an ESP32 microcontroller board. The ESP32 has built-in Wireless Fidelity (WiFi) + Bluetooth (BLE) and supports MicroPython. Also, machine learning algorithms such as Decision Trees (DTs), Support Vector Machines (SVMs), Random Forests (RFs), etc., are implemented using MicroPython and are deployed on a tiny edge computing device to monitor the activity of epileptic patients. All the adopted machine learning algorithms were compared in terms of performance metrics, such as Accuracy, Precision, Recall, False Alarm Rate (FAR), etc., and the efficacy of the device is analysed. The results demonstrate that the proposed device is capable of identifying activities of individuals, which is highly useful for epilepsy patients in monitoring their epileptic seizures. Furthermore, it is demonstrated that the proposed device is best deployed with an RF algorithm, since it exhibits an accuracy of 94.17%, which is better compared to that of the other machine learning algorithms. Also, the proposed device is simple and cost-effective and alerts caretakers of epilepsy patients with an FAR of less than 4%.

Plant disease prevention and management have become increasingly important due to population growth and the consequent intensification of crops, which drive the development of plant protection products. However, monitoring tools have also been integrated to improve food security while avoiding crop losses. Nowadays, early detection of pathogens like Phytophthora infestans in potatoes, Xylella fastidiosa in olives, and Fusarium species in cereals is sometimes the only viable alternative to developing targeted interventions. Implementing advanced technologies such as biosensors and artificial intelligence (AI) in agriculture can solve this problem, ensuring food security while protecting environmental health. Recent innovations in biosensor technology include smart sensors for real-time monitoring of soil conditions (pH, moisture, or total nutrient uptake), weather patterns, and crop/plant health, including the early detection of plant pathogens, herbicides, pesticides, heavy metals, and toxins.

This systematic review explores biosensors under the scope of precision agriculture (Agriculture 4.0) by integrating them with AI and the Internet of Things (IoT) to develop improved disease management strategies, increase crop yield, and optimize resources. Moreover, smartphone-based biosensors and machine learning (ML) algorithms further enhance the practicality of in-field applications through rapid data analysis and integration with precision agriculture systems.

The advantages, challenges, and knowledge gaps regarding the adoption of AI in biosensors and precision agriculture are also discussed. Future research should assess the effectiveness of these technologies in enhancing efficiency, productivity, and sustainability to enhance real-time decision making in agriculture.

The combination of CRISPR-based biosensors with smartphone technology is changing point-of-care (PoC) diagnosis by offering fast, portable, and highly sensitive diagnosis of the disease. CRISPR/Cas biosensing has recently been further developed with Cas12a, Cas13a, and Cas14 proteins, and new signal transduction strategies, including DNAzymes, binary 3D DNA walkers, and photoelectrochemical sensing have been introduced to boost the sensitivity and specificity of CRISPR diagnostics for real-life use. Moreover, electrochemical and field-effect transistor (FET)-based biosensors, quantum dot-assisted fluorescence, and surface-enhanced Raman scattering (SERS) have also been used to enhance the sensitivity and accuracy of the signal.

Lab-on-a-Chip (LoC) microfluidic platforms have been integrated with diagnostic devices to achieve automated and multiplexed testing for high-throughput and low-cost testing strategies. AI-based smartphone applications now offer real-time data analysis, cloud-based result sharing, and machine learning-based interpretations of results to enhance diagnostic care and decision-making. In addition to nucleic acid detection, CRISPR systems engineered from bacteria and viruses have been deployed to recognize proteins, metabolites, and small molecules for their use in infectious disease surveillance, cancer screening, antimicrobial resistance detection, and environmental monitoring. However, there are several problems, including reagent stability, large scale clinical validation, and efficient signal amplification. A major concern with respect to CRISPR reagents is their robustness in different environmental conditions, and this remains an issue that requires urgent solutions.

The main challenges include developing isothermal amplification methods and obtaining regulatory signatures. In the future, research should be aimed at the application of graphene-based nanomaterials for the increase in the signal transfer rate, the optimization of the CRISPR reagents to have a long shelf life, and the creation of wearable biosensors for healthcare purposes. The combination of CRISPR technology, nanomaterials, artificial intelligence, and digital health solutions is likely to redefine decentralized diagnostics and extend globally to affordable and scalable healthcare solutions.

Smartphone-based healthcare and diagnostic biosensors are adaptable, cost-effective, and accessible alternatives to traditional medical devices. This study examined smartphone-integrated biosensor advances in illness diagnosis, health monitoring, and personalized medicine. Biosensors on cell phones enable portable, real-time health examinations for a broad population by leveraging mobile technologies. How the electrochemical, optical, and biosensing biosensors detected biomarkers for diabetes, cardiovascular disease, infections, and cancer was also considered. This review focuses on user-friendly smartphone apps and mobile health (mHealth) platforms for data collection, processing, and interpretation. This study observed that even though smartphone-based biosensor technology has advanced, there are unresolved challenges. Addressing biosensor sensitivity and accuracy issues is crucial to reliable diagnostics. The effects of human variability, environmental disturbances, and sensor calibration on performance require improved sensor accuracy. Production costs, scaling problems, and healthcare system interactions slowed adoption in resource-limited regions. This study shows the need for affordable, easy-to-use sensors that can give patients and doctors vital information. Smartphone biosensors will benefit from nanotechnology, AI, and ML. These improvements should be smartphone biosensors' usability, precision, and efficiency, making them more reliable and versatile for consumer and medical use. Smartphone-based biosensors may improve individualized therapy, preventive care, and healthcare delivery, according to studies. This review also presents information on smartphone-based biosensor technology's advances, problems, and future uses.

Smartphone-based portable microscopic systems provide a viable alternative for meeting a variety of imaging and cell counting requirements in healthcare and other monitoring applications. Herein, we propose the design of a frugal microscopic imaging device that operates in bright-field mode using a smartphone for cell morphology and counting applications. Our device utilizes the inbuilt primary camera and the computational power of the phone. With the aid of readily available optical components, the designed platform is transformed into a high-throughput microscopic device. A custom-designed do-it-yourself (DIY) microfluidic chip, combined with a tailored Android app, simplifies the sample loading and automatic counting process. The microscopic device operates at three different optical magnifications and yields a lateral resolution of 1.21 µm over an acceptable field of view (FoV) of diameter ~4530 μm. The versatility of the system is demonstrated through imaging and counting of blood cells automatically. The results from our smartphone-based microscopic system show good agreement with values obtained from the widely used traditional hemocytometer. The affordability and portability of the proposed system suggest that it can be effectively implemented in resource-scarce areas. Additionally, we envision that the system would be significant for true point-of-care applications, research, and STEM education.

Smartphone-based colorimetric (bio)sensors are a promising alternative for developing affordable, deliverable, and user-friendly analytical tests for healthcare, food safety, and environmental monitoring. However, their effectiveness is limited by sensitivity to lighting conditions, which frequently requires the use of housings with controlled light sources that compromise affordability and simplicity. This study introduces a novel framework for enhancing smartphone-based colorimetric sensing via color space optimization. This approach enables accurate and consistent measurements under varying lighting conditions without additional housing. We evaluated the performance of smartphone-based colorimetric models to quantify monotonal color gradients with spectral compositions covering a wide range of visible spectra. In addition, we benchmarked the smartphone-based colorimetric models against absorbance-based models built using a benchtop UV-Vis spectrophotometer. Our findings indicate that smartphone-based quantification can achieve accuracy, precision, and detection limits comparable to absorbance-based models while offering a broader dynamic range. By assessing the quantification performance across several color spaces—RGB, HSV, and CIELAB—we found that the a* and b* chromatic coordinates of CIELAB demonstrate exceptional resilience to changes in illumination. We introduce the concept of Equichromatic Surfaces as an innovative framework for understanding the illumination resilience of CIELAB. This concept serves as a guide for developing reliable, housing-free, illumination-invariant optical (bio)sensors.

The present work addresses the need for a simple, cost-effective, and portable tool to detect specific nucleic acid sequences. Current nucleic acid detection methods are often complex and expensive, limiting their applicability in decentralized settings. The proposed approach utilizes a colorimetric, paper-based assay employing the sequence-specific cyanine dye DiSC₂(5) without the need for surface modification or sophisticated labelling systems. This portable device integrates a 3D-printed dark chamber with smartphone-based detection, allowing for image capture and analysis with the open-access ImageJ tool. A TATA box sequence, characterized by alternating A/T base pairs, was used as the model target in this study. The target was stored on a paper-based disk and subsequently detected through the blue color produced by the binding of DiSC₂(5). The assay effectively identifies TATA sequences, with the intensity of the blue color being directly correlated to target concentration. Specificity studies confirmed the assay's capability to differentiate between alternating and random A/T-containing sequences, highlighting its reliability. Furthermore, applications to synthetic PCR products demonstrate a limit of detection down to 10 nM. These results suggest that this portable, all-in-one, and low-cost platform could be effectively used for real-world samples, offering a promising alternative to traditional nucleic acid detection methods in point-of-care diagnostics and resource-limited settings.

Liquid biopsy has emerged as a transformative approach in modern diagnostics, offering a non-invasive means to detect and monitor cancer biomarkers such as microRNAs, circulating tumor DNA, and exosomes. Its versatility and potential for early detection have positioned it as a key player in advancing personalized medicine and real-time patient monitoring. However, challenges such as low biomarker concentrations and the need for accurate multiplexing persist. Lateral flow assays (LFAs) have evolved as versatile diagnostic tools, widely applied across diverse scientific disciplines. Recent advancements in artificial intelligence (AI) and automated image analysis have significantly enhanced the performance of LFAs, transforming them into user-friendly, point-of-care (POC) diagnostic devices. The integration of machine vision with LFAs represents a significant leap forward, enabling precise and real-time interpretation of results. This study introduces a novel multicolor LFA platform that leverages AI-driven image processing for the simultaneous detection and differentiation of three microRNA biomarkers (miR-21, miR-let-7a, and miR-155) in liquid biopsy applications. By employing distinct polystyrene beads as reporters, each color-coded to a specific microRNA, the system achieved multiplexed detection with limits as low as 1.56 fmol for each target. The innovative platform is paired with a smartphone-based application and a web application, which automate the reading and interpretation of test results, ensuring high accessibility and accuracy. The developed method was rigorously validated using real urine samples, demonstrating exceptional diagnostic performance with 99.3% accuracy, 99.1% sensitivity, and 100% specificity.

An array of gas sensors is combined with a mobile device to identify body odor. Metal-oxide semiconductor and nanomaterial-based sensors detect Volatile Organic Compounds (VOCs). VOCs like ammonia, acetic acid, trimethylamine, and hydrogen sulfide are associated with body odor. The array of sensors (MQ-135) captures the odor from the human body, and the system utilizes Artificial Intelligence (AI) algorithms to find the odor by using Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks for complex pattern recognition. The CNN layers identify the pattern across multiple sensors, i.e., spatial features. The spatial feature data become smoothened in the CNN layer and converted into a 1D vector. The LSTM receives this 1D vector as input to the model. The LSTM layers identify the odor intensity and composition over time. The MQ-135 is connected to the mobile device through a USB connection. This connection delivers real-time feedback to the user about the intensity of the odor like low, medium, or high. The user can then connect this device to a mobile device to identify human body odor. This procedure will not reduce the battery power of the mobile device. The proposed system is cost-efficient, portable, and accurate. It is important to focus on personal healthcare, hygiene, and wearable devices. In the future, gas sensors will be added to smart watches, sensor sensitivity will be increased, and better solutions can be provided by using AI models.

Portable real-time detection of Bisphenol A (BPA) in plastic-bottled water and large-scale milk samples presents significant challenges. Analytical techniques must prioritize specificity and sensitivity for accurate analysis. Affordability and rapid detection are also crucial to ensure practical application. Additionally, sustainability is essential when analyzing plastic-bottled water and large-scale water samples. The imaging sensor for smartphone-based portable CL imaging has significant potential for point-of-care applications. When combined with a smartphone readout device, it enables convenient and efficient environmental sensing, which makes it a promising tool for real-time detection in various environmental contexts. We designed and made the 96-well plate for CL image sensing of BPA by hydrothermally synthesizing a mixed-ligand MOF on the surface, characterized byspectrochemical characterization. With luminol, NaOH, H₂O₂, and BPA, the modified plate demonstrated high selectivity and tolerance for BPA. It exhibited distinct colour changes in the chemiluminescence (CL) image. Additionally, it was compatible with diluted plastic water and milk samples containing varying levels of BPA. Compared to conventional liquid chromatography–mass spectroscopy techniques, the modified plate yielded CL images at 1 pg/mL with recoveries exceeding 90% (n = 3), serving as a demand-driven alternative imaging device that eliminates the need for an additional light source and simplifies the basic operating system design essential for point-of-care biomedical diagnosis.