Artificial Intelligence: Powering Innovation in Biosensors and Waste Sorting

Artificial Intelligence (AI) is rapidly transforming industries, offering powerful solutions to complex challenges. While we often hear about AI in areas like self-driving cars or facial recognition, its impact is far broader, reaching into fields as diverse as healthcare diagnostics and environmental sustainability. Two examples from recent sources highlight AI's versatility: improving the accuracy of biosensors and enhancing the efficiency of municipal waste sorting.

Boosting Biosensor Accuracy with AI Individualization

Achieving high precision and reliability is crucial for biosensors used to measure specific analytes. One approach to improving this involves individualizing biosensors to a specific application scenario using artificial intelligence.

A method and system are proposed for measuring an analyte with a biosensor by enhancing its performance using high-performance parameters determined by artificial intelligence. The core process involves several steps:

Determining at least one quality parameter of the biosensor.
Determining at least one performance parameter and its associated threshold value based on at least one quality parameter using trained artificial intelligence.
Customizing the biosensor to a specific application scenario based on at least one performance parameter and its associated threshold value.

The biosensor typically includes a sensor and electronics, often with a test and reference cantilever. AI individualization means tailoring the biosensor's logic or parameters based on its specific characteristics and use case, ensuring reliable and consistent results.

The AI used in this process can be a trained neural network, using data such as time-resolved measurement curves and known analyte concentrations. The AI determines parameters like sensitivity, specificity, and yield, as well as quality parameters like the stability of a Wheatstone bridge's electrical resistance. By analyzing intrinsic properties, the system can compensate for variations during manufacturing or use, leading to more accurate and trustworthy measurements.

For more details, see the patent: WO2024209024A1

AI-Powered Robots Revolutionize Waste Sorting

AI is also proving transformative in sorting municipal solid waste (MSW). Traditional waste sorting plants, while highly automated, still rely on manual sorting for certain tasks, particularly for quality control and handling challenging waste streams like bulky waste. Manual sorting can involve low ergonomic conditions and decreasing performance over a work shift.

A research project tested the automation of municipal waste sorting plants using a robot with AI. The project, in a plant near Barcelona, used a ZRR2 robot system from ZenRobotics, aiming to supplement or replace manual sorting to increase recycling rates, improve purity, and enhance working conditions.

The ZRR2 robot system used two robot arms with RGB, NIR, and VIS sensors, plus a metal detector and deep learning software. The robot learned to recognize 13 different materials, including plastics, textiles, and Tetra Pak, through iterative training and feedback.

Empirical tests showed promising results: average purity of 90% (up to 100% in some cases) and later 97%, though recovery rates for some materials like textiles were lower. Challenges included the complex composition of municipal waste, gripping long or multi-layered items, and handling unstable inflow. Improvements included software updates and plant modifications.

For more details, see the open access paper: Artificial Intelligence in the Sorting of Municipal Waste as an Enabler of the Circular Economy [MDPI, 2021]

AI: A Versatile Tool for Progress

These two applications, though vastly different in scale and domain, showcase the power of artificial intelligence to enhance performance and address complex issues in real-world scenarios. In biosensing, AI enables the fine-tuning and individualization necessary for accurate and reliable measurements. In waste management, AI-powered robots offer a path to improved sorting efficiency, higher material purity, and better working conditions.

While challenges remain, such as optimizing recovery rates in waste sorting and addressing broader societal impacts, the successful implementation and testing described in these sources demonstrate AI's potential as a key enabler of progress in diverse fields, contributing to advancements in healthcare technology and the transition towards a more circular economy.

References:
Artificial Intelligence in the Sorting of Municipal Waste as an Enabler of the Circular Economy (MDPI, 2021)
WO2024209024A1: Method and System for Measuring an Analyte with a Biosensor (Patent PDF)

← Back to Blog