Meta-Learning Multi-Scale Radiology Medical Image Super-Resolution

High-resolution medical images have important medical value,but are difficult to obtain directly.Limited by hardware equipment and patient’s physical condition,the resolution of directly acquired medical images is often not high.Therefore,many researchers have thought of using super-resolution algorithms for secondary processing to obtain high-resolution medical images.However,current super-resolution algorithms only work on a single scale,and multiple networks need to be trained when super-resolution images of different scales are needed.This definitely raises the cost of acquiring high-resolution medical images.Thus,we propose a multi-scale superresolution algorithm using meta-learning.The algorithm combines a metalearning approach with an enhanced depth of residual super-resolution network to design a meta-upscale module.The meta-upscale module utilizes the weight prediction property of meta-learning and is able to perform the super-resolution task of medical images at any scale.Meanwhile,we design a non-integer mapping relation for super-resolution,which allows the network to be trained under non-integer magnification requirements.Compared to the state-of-the-art single-image super-resolution algorithm on computed tomography images of the pelvic region.The meta-learning multiscale superresolution algorithm obtained a surpassing of about 2%at a smaller model volume.Testing on different parts proves the high generalizability of our algorithm.Multi-scale super-resolution algorithms using meta-learning can compensate for hardware device defects and reduce secondary harm to patients while obtaining high-resolution medical images.It can be of great use in imaging related fields.

A Novel Unsupervised MRI Synthetic CT Image Generation Framework with Registration Network

In recent years,radiotherapy based only on Magnetic Resonance(MR)images has become a hot spot for radiotherapy planning research in the current medical field.However,functional computed tomography(CT)is still needed for dose calculation in the clinic.Recent deep-learning approaches to synthesized CT images from MR images have raised much research interest,making radiotherapy based only on MR images possible.In this paper,we proposed a novel unsupervised image synthesis framework with registration networks.This paper aims to enforce the constraints between the reconstructed image and the input image by registering the reconstructed image with the input image and registering the cycle-consistent image with the input image.Furthermore,this paper added ConvNeXt blocks to the network and used large kernel convolutional layers to improve the network’s ability to extract features.This research used the collected head and neck data of 180 patients with nasopharyngeal carcinoma to experiment and evaluate the training model with four evaluation metrics.At the same time,this research made a quantitative comparison of several commonly used model frameworks.We evaluate the model performance in four evaluation metrics which achieve Mean Absolute Error(MAE),Root Mean Square Error(RMSE),Peak Signal-to-Noise Ratio(PSNR),and Structural Similarity(SSIM)are 18.55±1.44,86.91±4.31,33.45±0.74 and 0.960±0.005,respectively.Compared with other methods,MAE decreased by 2.17,RMSE decreased by 7.82,PSNR increased by 0.76,and SSIM increased by 0.011.The results show that the model proposed in this paper outperforms other methods in the quality of image synthesis.The work in this paper is of guiding significance to the study of MR-only radiotherapy planning.

Dosimetric Comparison of Volumetric Modulated Arc Therapy (VMAT), 5F Intensity Modulated Radiotherapy (IMRT) and 3D Conformal Radiotherapy (3DCRT) in Rectal Carcinoma Receiving Neoadjuvant Chemoradiotherapy

Objective: To investigate better dosimetric distribution of volumetric modulated arc therapy (VMAT) vs. 5F intensity modulated radiotherapy (IMRT) and 3D conformal radiotherapy (3DCRT) in patients with locally advanced rectal cancer (LARC) when treated with neoadjuvant chemoradiotherapy. Methods: 3D-CRT, 5F-IMRT and VMAT plans for preoperative radiotherapy were 66011designed in 12 patients with locally advanced rectal cancer. The conformity index (CI) and homogeneity index (HI) in target volume, and the dose and volume of the organs at risk (OAR) irradiated including small bowel, bladder and bilatera1 femoral heads were compared among the three plans. Results: The CI for planning target volume (PTV) 2 and HI for PTV1 of VMRT and 5F-IMRT were superior to 3D-CRT. The CI of VMAT, 5F-IMRT and 3D-CRT plans were 0.71, 0.69 and 0.62 (p = 0.011 and p = 0.019, respectively). The HI of the VMAT and 5F-IMRT plans were both 1.04 and 3D-CRT planning was 1.06 (p = 0.022 and p = 0.006, respectively). The V35 - V45 of small bowel in VMAT were significantly less than in 5F-IMRT and 3D-CRT. V35 was 47.0, 56.4, and 72.8 cm3 for VMAT, 5F-IMRT, and 3D-CRT (p = 0.021 and p = 0.034, respectively), while V40 was 30.5, 35.5, 45.1 cm3 (p = 0.024 and p = 0.032, respectively) and V45 was 15.1, 18.1, 30.0 cm3 (p = 0.033 and p = 0.032, respectively). The D5, V30 and V50 of bladder in 3D-CRT were less than in VMAT and 5F-IMRT planning (p = 0.034, 0.004, 0.002 and p = 0.027, 0.003, 0.002, respectively). The Dmean of left femoral head in VMAT and 5F-IMRT were less than in 3D-CRT planning (p = 0.028 and p = 0.022, respectively) and the Dmean, V30 of right femoral head in VMAT and 5F-IMRT were better than in 3D-CRT planning (p = 0.044, 0.036 and p = 0.023, 0.028, respectively). Conclusions: Dosimetric analyses demonstrated that IMRT is superior to 3D-CRT in the conformity and homogeneity of dose distribution to the target volume, and provide a better protection to OARs sparing in patients with locally advanced rectal cancer for preoperative radiotherapy. With similar target coverage, VMRT is superior to 5F-IMRT in normal tissue sparing.

肠道膀胱准备下不同固定方式对前列腺癌精准放疗影响

目的比较3种体位固定装置在前列腺癌精准放疗中的摆位误差,为盆腔肿瘤精准放疗固定装置的选择和靶区外扩边界(MPTV)提供依据。方法回顾性分析中山大学肿瘤防治中心2015年4月至2020年12月133例需盆腔引流区照射的前列腺癌患者,采用1.2 m真空袋(39例)、1.8 m真空袋(44例)和本中心改进的个体化俯卧板(50例)固定。每位患者定位、放疗前均按流程进行肠道膀胱准备,每次治疗前锥形束CT与计划CT的配准采取相同配准框和算法,记录合格肠道膀胱的头脚、左右、腹背三个方向摆位误差,分析3种固定装置下3个方向摆位误差及相应MPTV数值,分析摆位误差与年龄、体重指数的相关性。结果3333组摆位误差数据得出,头脚、左右方向的1.2 m真空袋摆位误差均值(3.26、2.34 mm)均大于1.8 m真空袋(2.51、1.90 mm,P值均<0.001)和个体化俯卧板(3.07 mm,P=0.066;2.10 mm,P=0.009)。腹背方向的1.2 m真空袋(仰卧)摆位误差均值(2.20 mm)小于1.8 m真空袋(3.33 mm,P<0.001)和个体化俯卧板(3.61 mm,P<0.001)。1.8 m真空袋各方向摆位误差均小于个体化俯卧板(P≤0.028)。根据Van Herk外扩公式得出:1.2 m真空袋3个方向MPTV为4 mm左右,1.8 m真空袋和个体化俯卧板头脚、左右方向MPTV为3 mm左右,腹背方向>5 mm。摆位误差与年龄、BMI均不相关。结论摆位精准方面,1.8 m真空袋最优,个体化俯卧板次之;腹背方向仰卧位优于俯卧位。