Abstract
The diffuse optical tomography (DOT) technique which uses the traditional linear iterative algorithm has the problems of slow calculation speed and low reconstruction imaging accuracy in the inverse problem reconstruction, which limits its clinical application and development. This paper proposes an inverse-problem solving technology based on a stacked auto-encoder (SAE) network to improve the reconstruction accuracy of anomaly position, size and absorption coefficient in tissues. The reconstruction accuracy of anomaly position, size and absorption coefficient obtained by the traditional algebraic reconstruction technique (ART) method and the SAE based method are experimentally compared. The experimental results show that the SAE based method achieves the prediction accuracy of anomaly position of 96.25%, thus improving the accuracy and shortening the reconstruction time compared with the traditional ART method. Accordingly, the proposed method provides a better solution to the problem of the inaccurate reconstruction of the position and size of the rapid DOT based positioning of anomalies.
Highlights
The near-infrared spectroscopy for detection of biological tissue components has the advantages of excellent real-time performance, and continuous, non-invasive, and low-cost, characteristics
stacked auto-encoder (SAE) NEURAL NETWORK RECONSTRUCTION RESULTS To determine the anomaly position in the tissue using the method based on the SAE deep neural network, the radius of the anomaly was set to 10 mm, and the dataset of 430 samples was used for model training and testing, of which 350 samples constituted the training datasets, and 80 samples constituted the test datasets
This study proposes a reconstruction method for anomaly position based on the SAE neural network
Summary
The near-infrared spectroscopy for detection of biological tissue components has the advantages of excellent real-time performance, and continuous, non-invasive, and low-cost, characteristics. The diffuse optical tomography (DOT) technique [1] for detection of tissue components and imaging diagnosis of organisms has been widely applied in cerebral hematoma diagnosis [2], brain functional imaging [3], early screening and diagnosis of breast cancer [4], etc. Due to the rapid development of deep neural networks [9]–[11], they have been widely applied to various fields for feature extraction and classification of complex models. In [12], the neural network approach was adopted to reconstruct the optical parameters of biological tissues to overcome the shortcomings of traditional reconstruction methods such as a long calculation time and limited optical parameter range, providing the advantages of wide adaptability, and high accuracy and stability
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
