Positron Emission Tomography Lung Image Respiratory Motion Correcting with Equivariant Transformer

In addressing the challenge of motion artifacts in Positron Emission Tomography (PET) lung scans, our studyintroduces the Triple Equivariant Motion Transformer (TEMT), an innovative, unsupervised, deep-learningbasedframework for efficient respiratory motion correction in PET imaging. Unlike traditional techniques,which segment PET data into bins throughout a respiratory cycle and often face issues such as inefficiency andoveremphasis on certain artifacts, TEMT employs Convolutional Neural Networks (CNNs) for effective featureextraction and motion decomposition.TEMT’s unique approach involves transforming motion sequences into Liegroup domains to highlight fundamental motion patterns, coupled with employing competitive weighting forprecise target deformation field generation. Our empirical evaluations confirm TEMT’s superior performancein handling diverse PET lung datasets compared to existing image registration networks. Experimental resultsdemonstrate that TEMT achieved Dice indices of 91.40%, 85.41%, 79.78%, and 72.16% on simulated geometricphantom data, lung voxel phantom data, cardiopulmonary voxel phantom data, and clinical data, respectively. Tofacilitate further research and practical application, the TEMT framework, along with its implementation detailsand part of the simulation data, is made publicly accessible at https://github.com/yehaowei/temt.

Automatic Anomaly Detection of Respiratory Motion Based on Singular Spectrum Analysis

The realization of automatic anomaly detection of respiratory motion could be very useful to prevent accidental damage during radiation therapy. In this paper, we proposed an automatic anomaly detection method using singular value decomposition analysis. Before applying this method, the investigator needs a normal respiratory motion data of a patient. From these data, a trajectory matrix representing normal time-series feature is created. Decomposing the matrix, we obtained the feature of normal time series. Then, we applied the same procedure to real-time data and obtained real-time features. Calculating the similarity of those feature matrixes, an anomaly score was obtained. Patient motion was observed by a depth camera. In our simulation, two types of motion e.g. cough and sudden stop of breathing were successfully detected, while gradual change of respiratory cycle frequency was not detected clearly.

Impact of Field Strength and Respiratory Motion Control on Diffusion-Weighted MR Imaging of the Liver

Purpose: To evaluate the impact of field strength and respiratory motion control on diffusion-weighted MR imaging (DWI) of the liver at 1.5 and 3 T. Material and Methods: Three DWI sequences using seven b-values from 20 - 400 s/mm2 were designed with identical parameters but with different handling of respiratory motion [respiratory triggered (RT), free breathing (FB), breath hold (BH)] on 3 T and 1.5 T. Thirteen volunteers were examined at a 3 T and six of them also at a 1.5 T magnet. DW images were analyzed quantitatively and qualitatively. Regions of interest were placed in cranial, middle and caudal parts of the right liver lobe (RLL) and ADC and SNR were calculated. Results: ADC in RLL tended to be lower at 3 T MRI. Least inter-subject ADC variability was found with RT in the middle RLL at 3 T. Highest ADCs were found caudally in the RLL. Significant differences in ADC between middle and caudal RLL were calculated in FB and RT at 3 T and FB and BH at 1.5 T, respectively. No significant difference in SNR was found between 3 T and 1.5 T. There were significantly more artifacts in the left liver lobe (LLL) compared to the RLL in all sequences and in the LLL at 3 T compared to 1.5 T. Conclusion: Our study suggests that longitudinal hepatic ADC measurements should be performed using equivalent field strength, b-values, and acquisition technique, given influence of these factors on ADC measurements.

Dlung:Unsupervised Few-Shot Diffeomorphic Respiratory Motion Modeling

Lung image registration plays an important role in lung analysis applications,such as respiratory motion modeling.Unsupervised learning-based image registration methods that can compute the deformation without the requirement of supervision attract much attention.However,it is noteworthy that they have two drawbacks:they do not handle the problem of limited data and do not guarantee diffeomorphic(topologypreserving)properties,especially when large deformation exists in lung scans.In this paper,we present an unsupervised few-shot learning-based diffeomorphic lung image registration,namely Dlung.We employ fine-tuning techniques to solve the problem of limited data and apply the scaling and squaring method to accomplish the diffeomorphic registration.Furthermore,atlas-based registration on spatio-temporal(4D)images is performed and thoroughly compared with baseline methods.Dlung achieves the highest accuracy with diffeomorphic properties.It constructs accurate and fast respiratory motion models with limited data.This research extends our knowledge of respiratory motion modeling.

Fast-response,high-sensitivity multi-modal tactile sensors based on PPy/Ti_(3)C_(2)T_(x) films for multifunctional applications

In recent years,multi-modal flexible tactile sensors have become an important direction in the development of electronic skin because of their excellent sensitivity,flexibility and wearable properties.In this work,a humidity-pressure multi-modal flexible sensor based on polypyrrole(PPy)/Ti_(3)C_(2)T_(x) sensitive film packaged with porous polydimethylsiloxane(PDMS)is investigated by combining the sensitive structure generation mechanism of in situ polymerization to achieve the simultaneous detection of humidity and pressure,which has a sensitivity of 89,113.4Ω/%RH in a large humidity range of 0%-97%RH,and response/recovery time of 2.5/1.9 s.The tactile pressure sensing has a high sensitivity,a fast response of 67/52 ms,and a wide detection limit.The device also has excellent performance in terms of stability and repeatability,making it promising for respiratory pattern and motion detection.This work provides a new solution to address the construction of multi-modal tactile sensors with potential applications in the fields of medical health,epidemic prevention.