Posted in | News | Imaging

AI-Assisted Terahertz Imaging for Acute Lung Injury Therapy

In a study published in Biosensors, a label-free and rapid identification technique based on artificial intelligence (AI)-assisted terahertz imaging is proposed to identify the therapeutic effect of low-level laser therapy for acute lung injury.

Study: Artificial Intelligence-Assisted Terahertz Imaging for Rapid and Label-Free Identification of Efficient Light Formula in Laser Therapy. Image Credit: nito/Shutterstock.com

What is Acute Lung Injury?

Acute lung injury is a severe respiratory failure characterized by hypoxemia and the development of bilateral infiltrates on chest radiographs and is sometimes referred to as acute respiratory distress syndrome.

Acute lung injury continues to substantially cause high mortality and morbidity in critically sick patients.

Low-Level Laser Therapy Treatment for Acute Lung Injury

Due to the ineffectiveness of current therapies for acute lung injury, it is critical to develop an effective therapeutic approach to significantly lower mortality and enhance the life expectancy of patients.

Low-level laser therapy holds significant potential as a noninvasive therapeutic technique in biomedicine. The therapeutic use of low-level laser therapy was documented about 50 years ago. Since then, it has been an effective treatment for traumatic and inflammatory diseases with fewer side effects than similar therapies.

The therapeutic efficacy of low-level laser therapy is primarily determined by specific light formulae, including irradiation time and wavelength. However, current techniques for identifying effective light formulae are complex and time-consuming.

Therefore, developing label-free, highly sensitive, and rapid identification methodologies will significantly boost the advancement and use of low-level laser therapy.

Identifying the Therapeutic Effect of Low-Level Laser Therapy with Terahertz Imaging

Terahertz imaging has received a lot of interest as a potential radiological technique in biomedical applications. This is because it has high diagnostic safety and sensitivity and a unique IR fingerprint spectrum.

Terahertz imaging has investigated various disorders, such as skin burns, traumatic brain damage, arthritis, silkworm egg development phases, and various cancers. This makes it capable of quantifying cellular and tissue processes and tracking the development and remission of disease to assess the treatment efficacy.

However, hardware constraints in imaging systems restrict the use of terahertz imaging for biological diagnostics. For example, the major hindrance is the slow imaging speed resulting from the low imaging resolution and severe blurring.

Therefore, enhancing the sharpness and clarity of terahertz imaging while decreasing the imaging duration is essential for rapid image analysis.

Using AI Algorithms to Improve Identification Time of Terahertz Imaging

Automatic identification methodology based on a voting classifier and rapid super-resolution (SR) image reconstruction were implemented to design and improve the label-free terahertz imaging-based artificial intelligence identification method.

Using low-level laser therapy, the therapeutic effects of acute lung injury were determined by experimenting with various irradiation times and wavelengths. Vertical reflecting terahertz imaging systems and their gray histograms at 0.14 terahertz imaging were designed as the biophysical features to determine the treatment impact based on the sensitive terahertz spectrum response of acute lung injury.

The gray histograms were then processed with feature extraction algorithms, such as principal component analysis (PCA).

Lastly, the peak signal-to-noise ratio, mean square error, and structural similarity were applied to compare the performance of the four SR reconstruction algorithms implemented in rapid terahertz imaging: enhanced super-resolution generative adversarial network (ESRGAN), super-resolution generative adversarial network (SRGAN), enhanced deep residual network (EDSR), and residual channel attention network (RCAN).

Significant Findings of the Study

Although scanning terahertz imaging offers higher image resolution and needs only half the time of the systematic biological measurements in identifying light formulae, it is still not fast enough for real-time identification.

On the other hand, the modified terahertz camera imaging technique offers a rapid imaging rate, but its poor resolution restricts its biological application. Therefore, this study developed a super-resolution imaging reconstruction-based rapid terahertz imaging technology to minimize imaging time without compromising image resolution.

The proposed approach enables fast and label-free identification of the effective low-level laser therapy light formula for acute lung injury. It demonstrated 91.22% accuracy and 33 seconds identification time, 400 times faster than the biological method and over 200 times faster than the scanning terahertz imaging technique.

The terahertz spectral response of acute lung injury revealed an impressive absorption peak in the 0.1-0.15 THz region.

AI-assisted terahertz imaging was solely employed to identify biological tissue samples encased in paraffin. However, it could be applied to develop and enhance other biological techniques, such as photodynamic therapy.

The AI-assisted terahertz imaging can be improved by integrating super-resolution and compressed sensing algorithms.

Reference

Shi, J., Guo, Z., Chen, H., Xiao, Z., Bai, H., Li, X., Niu, P., & Yao, J. (2022) Artificial Intelligence-Assisted Terahertz Imaging for Rapid and Label-Free Identification of Efficient Light Formula in Laser Therapy. Biosensors. https://www.mdpi.com/2079-6374/12/10/826

Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.

Owais Ali

Written by

Owais Ali

NEBOSH certified Mechanical Engineer with 3 years of experience as a technical writer and editor. Owais is interested in occupational health and safety, computer hardware, industrial and mobile robotics. During his academic career, Owais worked on several research projects regarding mobile robots, notably the Autonomous Fire Fighting Mobile Robot. The designed mobile robot could navigate, detect and extinguish fire autonomously. Arduino Uno was used as the microcontroller to control the flame sensors' input and output of the flame extinguisher. Apart from his professional life, Owais is an avid book reader and a huge computer technology enthusiast and likes to keep himself updated regarding developments in the computer industry.

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Ali, Owais. (2022, October 07). AI-Assisted Terahertz Imaging for Acute Lung Injury Therapy. AZoOptics. Retrieved on November 28, 2022 from https://www.azooptics.com/News.aspx?newsID=28006.

  • MLA

    Ali, Owais. "AI-Assisted Terahertz Imaging for Acute Lung Injury Therapy". AZoOptics. 28 November 2022. <https://www.azooptics.com/News.aspx?newsID=28006>.

  • Chicago

    Ali, Owais. "AI-Assisted Terahertz Imaging for Acute Lung Injury Therapy". AZoOptics. https://www.azooptics.com/News.aspx?newsID=28006. (accessed November 28, 2022).

  • Harvard

    Ali, Owais. 2022. AI-Assisted Terahertz Imaging for Acute Lung Injury Therapy. AZoOptics, viewed 28 November 2022, https://www.azooptics.com/News.aspx?newsID=28006.

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit