Detection of airborne nanoparticles with lateral shearing digital holographic microscopy. (April 2022)

• Record Type: Journal Article
• Authors: Ustabas Kaya, Gulhan; Kocabas, Sefa; Kartal, Seda; Kaya, Hakan; Tekin, Ishak Ozel; Tigli Aydin, Rahime Seda; Kutoglu, Senol Hakan
• Is Part Of: Optics and lasers in engineering. Volume 151(2022)
• Publication Date: 2022-04
• Subjects: Nano/micro-sized particles -- Concentration -- Detection -- Lateral shearing digital holography -- Classification -- Deep learning classifier; Lasers in engineering -- Periodicals; Optical measurements -- Periodicals; Optics -- Periodicals; Lasers en ingénierie -- Périodiques; Mesures optiques -- Périodiques; Optique -- Périodiques; 621.36605
• DOI: 10.1016/j.optlaseng.2021.106934
• Languages: English
• ISSNs: 0143-8166

Abstract

Highlights: Lateral shearing digital holographic microscopy (LS-DHM) is used for detection. Detection of the airborne nano/microsized particles via LS-DHM. Observation of the airborne particle concentration at different humidity values. The type and size of the particles are determined with deep learning classifier. Abstract: In this study, we propose to image the nano-sized particles dispersed in the air with the lateral shearing digital holography technique. Another aim of ours is to determine how long these particles stay in the environment (cabinet, room, air and etc.) by calculating the concentration amounts (CA) in different humidity conditions and what type of particle is given to the cabinet. Binarized phase images reconstructed from the recorded holograms were used to find the CA. The CA was calculated by measuring in a time depended manner. In addition, in this study, the behavior of particulate matter in different humidity conditions was also revealed. It has been clearly demonstrated that the residence time in the air varies depending on the structure of each particulate matter and the behavior of the environment in different humidity conditions. Moreover, images were classified by deep learning based on Convolution Neural Network (CCN) algorithm. The CNN algorithm was trained using binarized phase images and the type of substance dispersed in the air was determined with high accuracy (99%). Unlike the traditional digital holography microscopy techniques used to capture a single microsized sample located between two lamellas, this study reveals successful results obtained with high accuracy that nano-sized particles moving freely in the air can be imaged and the type of particulate matter can be detected. It is thought that the proposed system for the early detection of particles that causes various infectious diseases will lead to future studies in terms of both non-contact and real-time imaging.