摘要
一位新闻记者兼机器人与机器学习的工作人员新闻编辑每日新闻-人工智能的新研究是一篇报道的主题。根据NewsRx记者在中华人民共和国广州的新闻报道，研究称：“本研究旨在开发人工智能模型，预测(RRD)孔源性视网膜脱离患者的术后功能转归。对184例确诊为RR D的患者进行了前瞻性回顾和数据提取。”新闻记者引用了南方医科大学的一项研究:“主要结果是术后3个月矫正视力最佳(BCVA),BCVA小于6/18 Snellen Acity的患者被分为视力障碍组。采用术前预测因素建立了深度学习模型,包括超宽视野眼底图像、眼底图像和眼底图像。”结果:(OCT)黄斑区、年龄、性别、术前BCVA的图像,采用融合的方法捕捉不同模式在模型构建过程中的相互作用,74例(40%)患者治疗后仍有视力障碍,年龄、性别、术前BCVA、眼压、黄斑脱离、视网膜病变、视网膜多模式融合模型在(AUC)曲线下的平均面积为0.91，平均精度为0.86，敏感性为0.94，特异性为0.80.，表明黄斑区是最活跃的区域。正如在OCT和Ultra-Widefiel D图像中观察到的那样。这项初步研究表明，人工智能技术在预测RRD手术后功能结果方面可以获得较高的AUC，即EVE N。机器学习方法确定黄斑区是最活跃的区域。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – New research on Artificial Intelligenc e is the subject of a report. According to news reporting from Guangzhou, People ’s Republic of China, by NewsRx journalists, research stated, “This study aimed to develop artificial intelligence models for predicting postoperative functiona l outcomes in patients with rhegmatogenous retinal detachment (RRD). A retrospec tive review and data extraction were conducted on 184 patients diagnosed with RR D who underwent pars plana vitrectomy (PPV) and gas tamponade.” The news correspondents obtained a quote from the research from Southern Medical University, “The primary outcome was the best-corrected visual acuity (BCVA) at three months after the surgery. Those with a BCVA of less than 6/18 Snellen acu ity were classified into a vision impairment group. A deep learning model was de veloped using presurgical predictors, including ultra-widefield fundus images, s tructural optical coherence tomography (OCT) images of the macular region, age, gender, and preoperative BCVA. A fusion method was used to capture the interacti on between different modalities during model construction. Among the participant s, 74 (40%) still had vision impairment after the treatment. There were significant differences in age, gender, presurgical BCVA, intraocular press ure, macular detachment, and extension of retinal detachment between the vision impairment and vision non-impairment groups. The multimodal fusion model achieve d a mean area under the curve (AUC) of 0.91, with a mean accuracy of 0.86, sensi tivity of 0.94, and specificity of 0.80. Heatmaps revealed that the macular invo lvement was the most active area, as observed in both the OCT and ultra-widefiel d images. This pilot study demonstrates that artificial intelligence techniques can achieve a high AUC for predicting functional outcomes after RRD surgery, eve n with a small sample size. Machine learning methods identified The macular regi on as the most active region.”
NETL
NSTL