Diagnostic Performance of Ultrasound Computer-Aided Diagnosis Software Compared with That of Radiologists with Different Levels of Expertise for Thyroid Malignancy: A Multicenter Prospective Study

The aim of the work described here was to evaluate the diagnostic performance of ultrasound thyroid computer-aided diagnosis (CAD) software. This multicenter prospective study included 494 patients (565 thyroid nodules) who underwent surgery or biopsy after ultrasonography at four hospitals from January 2019 to September 2019. The diagnostic performance metrics of different readers were calculated and compared with the pathologic results. The sensitivity of CAD was outstanding and was equivalent to that of a senior radiologist (90.51% vs. 88.47%, p > 0.05). The area under the curve of CAD was equivalent to that of a junior radiologist (0.748 vs. 0.739, p > 0.05). However, the specificity was only 49.63%, which was lower than those of the three radiologists (75.56%, 85.93% and 90.37% for the junior, intermediate and senior radiologists, respectively). The diagnostic performance of the junior radiologist was significantly improved with the aid of CAD (junior + CAD). The sensitivity and area under the curve of junior + CAD were improved from 72.20% to 89.93% and from 0.739 to 0.816, respectively (both p values <0.05), and the positive predictive value, negative predictive value and κ coefficient improved from 76.3% to 78.6%, 82.0% to 86.8% and 0.394 to 0.511, respectively. Though specificity slightly decreased from 75.56% to 73.33%, the difference was not statistically significant (p > 0.05). In general, the clinical application value of CAD is promising, and its instrumental value for junior radiologists is significant.     

计算机辅助诊断系统对甲状腺结节的诊断价值研究

目的比较计算机辅助诊断（CAD）系统与多名超声医师对甲状腺结节的诊断效能，初步探讨CAD软件的诊断价值及分析甲状腺结节超声特征对CAD软件及超声医师诊断的影响。 方法选取2016年2月至2018年6月电子科技大学医学院附属四川省肿瘤医院医学影像信息（PACS）系统中甲状腺结节灰阶超声图像50张，采用CAD软件及111名超声医师同时对50张甲状腺结节图像进行诊断。以病理结果为"金标准"分别计算CAD软件，准确率最高的高年资医师、准确率最高的低年资医师鉴别诊断甲状腺结节的敏感度、特异度、阳性预测值、阴性预测值，并绘制受试者工作特征曲线（ROC曲线）；各组间准确率的比较采用McNemar检验，ROC曲线下面积的比较采用Z检验。 结果CAD软件、准确率最高的高年资医师、准确率最高的低年资医师诊断甲状腺结节良恶性的敏感度、特异度、阳性预测值、阴性预测值及准确性分别为76.9%、87.5%、86.9%、77.8%、82.0%；86.9%、77.8%、76.9%、87.5%、82.0%；82.6%、70.4%、70.4%、82.6%、76%；CAD软件与高年资医师诊断准确率相同且均高于低年资医师，差异有统计学意义；CAD软件与高年资医师ROC曲线下面积一致且均大于低年资医师，但差异均无统计学意义（P均>0.05）。医师误诊的病例主要为桥本甲状腺炎以及微小低回声病灶伴点状强回声的甲状腺结节，而分布位置以及结节内粗大钙化灶伴后方宽大声影造成了CAD软件的误诊。 结论CAD软件诊断甲状腺结节的准确率与高年资医师一致，高于低年资医师；甲状腺结节的分布位置以及结节内粗大钙化灶伴后方宽大声影可能是影响CAD软件诊断准确性的因素；而桥本甲状腺炎以及微小低回声病灶伴点状强回声可能会影响超声医师对甲状腺结节的正确诊断。     

计算机辅助诊断系统在甲状腺结节超声诊断中的应用

[摘 要] 目的 评估计算机辅助诊断（Computer-aided Diagnosis ,CAD）系统在甲状腺超声检查中的诊断效能及临床应用价值。方法 选取2018年8月至2019年1月在中南大学湘雅三医院进行甲状腺超声检查并行手术切除的171例患者，共205个甲状腺结节。分别采用CAD及4名不同经验水平的超声医师对205例甲状腺结节的超声图像进行分析并根据美国放射学会（ACR）的TI-RADS指南进行分类，然后对4名不同经验水平的超声医师结合CAD也进行了研究。以手术病理结果为金标准,评估CAD系统在鉴别甲状腺良恶性结节中的诊断效能，以及CAD对不同经验水平的超声医师的影响。结果 CAD系统的使用提高了4名超声医师对甲状腺结节鉴别诊断的敏感性及AUC值(结合CAD系统vs未结合CAD系统:灵敏度：超声医师A, 93.10% vs 87.93%;超声医师B, 90.52% vs 84.48%;超声医师C, 85.34% vs 78.45%;超声医师D, 75.00% vs 66.38%，AUC值：超声医师A, 0.95 vs 0.94;超声医师B, 0.93 vs 0.92;超声医师C, 0.86vs 0.81;超声医师D, 0.86 vs 0.70)，差异均有统计学意义（P均＜0.05）。然而CAD系统的低特异性(73.03%)仅对超声医师C、D有显著性改善(P均＜0.05)。CAD系统对甲状腺恶性肿瘤的诊断敏感性与具有5年经验的超声医师相似，差异无统计学意义（P=1.00），但CAD系统的特异性较低。结论 CAD鉴别诊断甲状腺结节的敏感性较高，但特异性较低。结合CAD可有效提高初级医师甲状腺结节的超声诊断水平，也能提高高年资医师对甲状腺结节鉴别诊断的敏感性。     

超声诊断喉癌的准确性及诊断医师之间的一致性研究

目的研究超声诊断喉癌的准确性及诊断医师之间的一致性。方法对121例临床疑似喉癌的病例进行超声检查并采集图像,再分别由3位超声医师根据采集到的图像和喉癌超声诊断标准分级诊断为Ⅰ～Ⅵ级。对照病理结果后构建ROC曲线,计算曲线下面积(Az)以评价诊断喉癌的准确性。分别计算κ值以评价3位医师之间的一致性。结果 118例患者获得了满意的超声图像。3位医师对应的Az值分别为0.939、0.954和0.947,各医师之间的κ值分别为0.75(P0.001)、0.71(P0.001)、和0.86(P0.001)。结论超声诊断喉癌的准确性和一致性较高,在喉癌的影像诊断方面有着良好的应用前景。     

数字胸片多频域后处理对计算机辅助检测系统肺结节检测的影响

目的评价数字胸片多频域后处理对计算机辅助检测(CAD)系统肺结节检出的影响。方法对经CT证实的54例肺结节病例及54例正常者的数字胸片选择三种不同的structure preference参数进行多频域后处理,在1次曝光的条件下各得到标准图像、高通过图像及低通过图像三组图像,然后,采用计算机辅助检测系统(IQQATM-Chest V1.2)进行阅片,首先由2名观察者共同根据CT结果客观分析并记录CAD检出的结节数及假阳性数。其次,由另外4名观察者(高年资及低年资放射医师各2名)应用CAD系统所提供的肺结节的智能质量和数量分析功能,对CAD输出图像独立进行分析,并记录检出结节数及假阳性数,采用受试者操作特征(ROC)曲线分析观察结果。结果三组图像中,低通过组图像的平均ROC曲线下面积最大,人-机交互前、高年资及低年资人-机交互后的平均ROC曲线下面积分别为0.77、0.81、0.80。高通过组图像的平均ROC曲线下面积为最小,人-机交互前、高年资及低年资人-机交互后的平均ROC曲线下面积分别为0.57、0.67、0.71。三组图像的平均ROC曲线下面积存在差异(P<0.01)。结论数字胸片多频域后处理影响计算机辅助检测系统肺结节的检出。     

Effect of a novel segmentation algorithm on radiologists' diagnosis of breast masses using ultrasound imaging

We investigated the effect of using a novel segmentation algorithm on radiologists’ sensitivity and specificity for discriminating malignant masses from benign masses using ultrasound. Five-hundred ten conventional ultrasound images were processed by a novel segmentation algorithm. Five radiologists were invited to analyze the original and computerized images independently. Performances of radiologists with or without computer aid were evaluated by receiver operating characteristic (ROC) curve analysis. The masses became more obvious after being processed by the segmentation algorithm. Without using the algorithm, the areas under the ROC curve (Az) of the five radiologists ranged from 0.70∼0.84. Using the algorithm, the Az increased significantly (range, 0.79∼0.88; p < 0.001). The proposed segmentation algorithm could improve the radiologists’ diagnosis performance by reducing the image speckles and extracting the mass margin characteristics.     

基于计算机辅助诊断技术的超声图像处理软件对甲状腺结节诊断效能的初步研究

目的通过比较基于计算机辅助诊断技术（CAD）的甲状腺超声图像处理软件（安克侦）与超声医师对甲状腺结节进行甲状腺影像报告和数据系统（TI-RADS）评分的差别,初步探讨安克侦CAD软件在甲状腺结节良恶性鉴别诊断中的价值。 方法选取2015年5月至2016年10月就诊于中山大学附属肿瘤医院的194例甲状腺结节患者的病例资料。每个病例选取一个结节进行分析,所有结节均经手术或超声引导下细针穿刺活检（US-FNA）病理证实。由两位年资不同的超声医师评估结节图像,分别对结节的5个特征（成分、回声、形态、边缘、强回声）进行评分,得出总的TI-RADS评分并分类。安克侦CAD软件则通过自动分析结节图像,得出基于美国放射学会（ACR）标准的TI-RADS评分。 结果安克侦CAD软件的TI-RADS评分与高年资医师比较,差异无统计学意义（Z=0.964,P=0.335）,与低年资医师比较,差异具有统计学意义（Z=5.593,P<0.001）。以TR5为恶性诊断界值,安克侦CAD软件的诊断敏感度与高年资医师比较,差异无统计学意义（84.62% vs 86.54%,P=0.815）,且其略高于低年资医师,但差异无统计学意义（84.62% vs 73.08%,P=0.052）。安克侦CAD软件、低年资医师、高年资医师评分的诊断特异度分别为65.56%、87.78%、82.22%,安克侦CAD软件明显低于低年资和高年资医师（65.56% vs 87.78%,χ²=12.893,P<0.01;65.56% vs 82.22%,P=0.004）。CAD软件的ROC曲线下面积均低于高年资及低年资医师（0.735 vs 0.921,Z=4.537,P<0.0001;0.735 vs 0.898,Z=4.033,P=0.0001）。 结论安克侦CAD软件对甲状腺结节的诊断敏感度与高年资及低年资医师相比,并无显著差异,但诊断特异度及诊断准确性均低于高年资及低年资医师。其综合诊断效能仍需进一步深入研究。     

Reliability of 3-D Virtual Abdominal Tele-ultrasonography in Pediatric Emergency: Comparison with Standard-of-Care Ultrasound Examination

Ultrasound is currently recommended as the first-line examination for abdominal symptoms in children. However, a pediatric radiologist is not always available on site, especially during on-call duty. This study was aimed at evaluating the reliability of an innovative 3-D virtual abdominal tele-ultrasonography in this context. A prospective study was conducted between December 2020 and May 2021 that recruited 103 children undergoing ultrasound for abdominal pain. Trauma cases were excluded. Four tridimensional acquisitions were performed with a Smart Sensor 3D device (Canon Medical Systems, Otawara, Japan). Each tele-ultrasonography was secondarily blindly reviewed by two radiologists (one senior and one resident) with Fusion software (Canon Medical Systems). Acceptance and quality of the acquisitions were evaluated on a Likert scale. Inter-rater reliability was quantified using Cohen's κ coefficient and intraclass correlation coefficient. The ultrasound examination was normal in 66 cases (64%), abnormal in 36 cases (35%) and inconclusive in 1 case (1%). The acquisitions were obtained without objections from the children, their parents or the operators in more than 95% of cases. The quality of the acquisitions was considered good to excellent in 84% and 70% of cases. The sensitivity of the senior radiologist and the resident was 86% and 84%, respectively; specificity was 95% and 92%, positive predictive value 92% and 86% and negative predictive value 92 and 91% when comparing the conclusions of the standard and the tele-ultrasound examinations. Cohen's κ coefficients of the diagnosis obtained with the standard and the tele-ultrasound examinations were 0.82 and 0.71, respectively. The inter-rater Cohen's κ coefficient was 0.84. The intraclass correlation coefficient between the standard abdominal examination and the 3-D tele-ultrasound reformatted images for the following quantitative variables on pathological cases was 0.99 (confidence interval: 0.98–0.99). Virtual abdominal tele-ultrasonography is a promising method in pediatric emergencies.     

超声造影对肝复杂囊性病变的诊断性能和一致性研究

目的 评估超声造影诊断肝复杂囊性病变的应用价值.方法 65例患者中的67个肝复杂囊性病变接受了普通超声和超声造影检查,两位超声医师(一位年资高者和一位年资低者)独立阅片,应用ROC曲线评估普通超声和超声造影判断肝复杂囊性病变良恶性的诊断价值,Kappa值检验评估年资高及年资低者之间的一致性.结果 ROC曲线显示年资高者和年资低者得到的曲线下面积在普通超声分别为0.917和0.774(P=0.044),在超声造影分别为0.935和0.922(P=0.420).普通超声和超声造影的曲线下面积在年资高者差异无统计学意义(0.917对0.935,P=0.380),在年资低者差异有统计学意义(0.774对0.922,P=0.047).超声造影明显提高了诊断肝复杂囊性病变良恶性的特异性、准确性和阳性预测值(均P＜0.01).普通超声和超声造影对具体病变的正确诊断率在年资高者分别为26.9%(18/67)和76.1%(51/67),在年资低者分别为28.4%(19/67)和58.2%(39/67)(均P＜0.001).受试者之间的一致性超声造影较普通超声高,Kappa值分别为0.774(95%置信区间:0.688～0.860)和0.325(95%置信区间:0.214～0.436).结论与普通超声相比,超声造影显著提高了鉴别诊断肝复杂囊性病变良恶性的能力和对具体病变的诊断能力,并且提高了观察者间的一致性.     

基于甲状腺成像报告和数据系统分类的计算机辅助诊断系统对甲状腺癌超声诊断的辅助价值

目的探讨基于甲状腺成像报告和数据系统（TI-RADS）分类的计算机辅助诊断（CAD）系统对超声医师诊断甲状腺癌的辅助价值。 方法收集2018年10月至2019年3月在国内5家医院的400例甲状腺结节超声图像进行多中心回顾性研究。采用由北京大学前沿交叉学科研究院研发的基于TI-RADS分类的CAD系统,超声医师诊断甲状腺癌的诊断模式分为无CAD模式和CAD模式,11名具有不同工作经验的超声医师（低年资超声医师4名,中年资超声医师4名,高年资超声医师3名）在上述2种模式下诊断甲状腺癌。比较2种诊断模式的诊断效能及读片时间：绘制CAD系统和超声医师诊断甲状腺癌的受试者工作特征（ROC）曲线,应用DeLong方法比较2种模式下曲线下面积（AUC）的差异;应用配对t检验比较2种模式的读片时间差异。 结果在CAD模式下,所有超声医师诊断甲状腺癌的AUC值较无CAD模式有显著提高［0.848（0.837~0.858） vs 0.800（0.788~0.812）］,差异具有统计学意义（P＜0.001）;敏感度从73.8%（95%CI：71.9%~75.6%）提升到82.7%（95%CI：81.0%~84.3%）,差异具有统计学意义（χ²=9.870,P＜0.001）;特异度从86.2%（84.7%~87.6%）提升到86.9%（85.4%~88.3%）,但差异无统计学意义（χ²=0.021,P=0.379）。进行亚组分析时,在CAD模式下,低年资超声医师和中年资超声医师诊断甲状腺癌的AUC值较无CAD模式有显著提高（0.840 vs 0.740;0.848 vs 0.814）,差异具有统计学意义（P＜0.001、=0.001）,但高年资超声医师诊断甲状腺癌的AUC值较无CAD模式无明显变化（0.859 vs 0.861,P=0.861）。在CAD模式下,所有超声医师的阅片时间较无CAD模式减少［（20.2±8.2）s vs（22.7±9.6）s］,差异具有统计学意义（t=-23.9,P＜0.001）。 结论CAD模式有助于低年资超声医师和中年资超声医师对甲状腺癌的诊断,同时缩短了诊断时间。     

计算机辅助检测和诊断系统对甲状腺结节细针穿刺活检的评估价值

目的探讨计算机辅助检测和诊断（CAD）系统评估甲状腺结节是否应行细针穿刺活检的临床应用价值。 方法收集2019年2月至10月哈尔滨医科大学附属第一医院106例欲行细针穿刺活检的甲状腺结节,进行CAD系统、高年资、低年资超声医师评估并按照美国甲状腺学会指南建议穿刺与否,以细针穿刺活检病理结果为“金标准”进行判读。若病理结果为恶性,则判定建议细针穿刺活检有意义;若病理结果为良性,则判定建议细针穿刺活检无意义,分别计算3种方式的敏感度、特异度、阳性预测值、阴性预测值和准确性,比较3种方式的评估效能;按照结节大小将其分为3组：1.0 cm~＜1.5 cm组、1.5 cm~＜2.0 cm组和≥2.0 cm组,研究其对CAD软件评估甲状腺结节穿刺与否这一效能的影响,比较不同组的敏感度、特异度和准确性的差异。3种方式间及各组间差异的比较均采用McNemar检验法。 结果CAD软件的评估效能与高年资医师相对比,CAD软件的敏感度低,而特异度高,差异均有统计学意义（71.2% vs 84.6%,P=0.039;77.8% vs 61.1%,P=0.049）;CAD软件的评估效能与低年资医师相对比,其敏感度、特异度均高于低年资医师,差异均有统计学意义（71.2% vs 55.8%,P=0.021;77.8% vs 64.8%,P=0.039）;CAD软件的评估准确性与高年资医师相近,但大于低年资医师（74.5% vs 60.4%）,差异有统计学意义（P=0.001）。CAD软的评估效能与结节大小有关,当结节最大直径为1.0 cm~＜1.5 cm时,CAD软件的敏感度最高（82.4% vs 40.0% vs 62.5%）,差异具有统计学意义（P=0.027）;当结节最大直径≥2.0 cm时,CAD软件具有最高的特异度及准确性（91.7% vs 80.0% vs 53.3%;84.4% vs 81.6% vs 48.0%）,差异均具有统计学意义（P=0.023、0.002）。 结论CAD系统能对甲状腺结节给出合理的细针穿刺活检建议,其评估准确性与高年资医师相近且高于低年资医师,结节大小能影响其评估效能。     

超声造影与常规超声鉴别诊断附件区肿块的比较研究

目的 比较常规超声和实时超声造影对附件区肿块的鉴别诊断性能.方法 对137例附件区肿块行常规超声和实时超声造影检查.两位超声医师(受试者)对检查结果独立作出诊断,用ROC曲线评估常规超声和实时超声造影对附件区肿块的鉴别诊断性能,Kappa值评估受试者之间的一致性.结果 ROC曲线显示两位受试者得到的曲线下面积:常规超声分别为0.731和0.738,超声造影分别为0.891和0.903(P＜0.01).超声造影能明显提高判别病变良恶性的准确性.受试者之间的一致性超声造影高于常规超声,κ值分别为0.893和0.681.两位受试者对附件区病变的诊断正确率常规超声分别为73.2％和74.2％,超声造影分别为90.7％和91.2％(均P＜0.01).结论 实时超声造影可明显提高超声对附件区病变的鉴别诊断能力.     

CAD(计算机辅助探察)对提高乳腺癌诊断敏感性的意义

目的探讨CAD技术对提高乳腺癌诊断敏感性的意义.方法回顾性分析49位患者的常规乳腺钼靶照片(共162张),分别由低年、中年临床医师、CAD独立阅片及低年医师 CAD共同读片,将结果与最终诊断结果相比较.结果恶性钙化19例:低年医师漏诊7例;中年医师漏诊1例;CAD漏诊1例;低年医师 CAD漏诊2例.恶性肿块14例:低年医师漏诊6例;中年医师漏诊2例;CAD漏诊6例;低年医师 CAD漏诊3例.结论 CAD可以提高乳腺癌诊断的敏感性.     

Inter-professional agreement of ultrasound-based diagnoses in patients with shoulder pain between physical therapists and radiologists in the Netherlands

Study designReliability study.ObjectivesThe aim of this study was to evaluate the interrater-reliability of the interpretation of diagnostic ultrasound in patients with shoulder pain between physical therapists and radiologists.BackgroundAlthough physical therapists in The Netherlands increasingly use diagnostic ultrasound in clinical practice, there is no evidence available on its reliability.MethodsA cohort study included patients with shoulder pain from primary care physiotherapy. Patients followed the usual diagnostic pathway of which diagnostic ultrasound could be a part. Patients that received diagnostic ultrasound also visited a radiologist within one week for a second one. Patients and radiologists were blinded for the diagnostic ultrasound diagnosis of the physical therapists. Agreement was assessed using Cohen's kappa statistics. Subgroup analysis was performed on education and experience.ResultsA total of 65 patients were enrolled and 13 physical therapists and 9 radiologists performed diagnostic ultrasound. We found substantial agreement (0.63 K) between physical therapists and radiologists on the assessment of full thickness tears. The overall kappa of all four diagnostic categories was 0.36, indicating fair agreement. The more experienced and highly trained physical therapists showed moderate agreement (0.43 K) compared to only slight agreement (0.17 and 0.09 K) from the less experienced and trained physical therapists with radiologists.ConclusionThe reliability between physical therapists and radiologist on diagnostic ultrasound of shoulder patients in primary care is borderline substantial (Kappa = 0.63) for full thickness tears only. This level of reliability is relatively low when compared with the high reliability between radiologists. More experience and training of physical therapists may increase the reliability of diagnostic ultrasound.     

Computerized analysis of images in the detection and diagnosis of breast cancer.

Improvements in mammographic acquisition techniques have resulted in making the early signs of breast cancer more apparent on mammograms. However, the accuracy of the overall mammographic examination depends on both the quality of the mammographic images and the ability of the radiologist to interpret those images. While mammography is the best screening method for the early detection of breast cancer, radiologists do miss lesions on mammograms. Use of output, however, from a computerized analysis of an image by a radiologist may help him/her in the detection or diagnostic tasks, and potentially improve the overall interpretation of breast images and the subsequent patient care. Computer-aided detection and diagnosis (CAD) involves the application of computer technology to the process of medical image interpretation. CAD can be defined as a diagnosis made by a radiologist, who uses the output from a computerized analysis of medical images as a "second opinion" in detecting and diagnosing lesions, with the final diagnosis being made by the radiologist. The computer output must be at a sufficient performance level, and in addition, the output must be displayed in a user-friendly format for effective and efficient use by the radiologist. This chapter reviews CAD in breast cancer detection and diagnosis, including examples of image analyses, multi-modality approaches (i.e., special-view diagnostic mammography, ultrasound, and MRI), and means of communicating the computer output to the human.     

Principal component regression-based contrast-enhanced ultrasound evaluation system for the management of BI-RADS US 4A breast masses: objective assistance for radiologists

A portion of detected breast masses might be overrated by using the Breast Imaging-Reporting and Data System ultrasonography (BI-RADS US) lexicon. A principal component regression-based contrast-enhanced ultrasound (PCR-CEUS) evaluation system was built to quantitatively illustrate whether CEUS could help radiologists to differentiate 4A masses. The PCR-CEUS evaluation system, based on principal component analysis (PCA) and logistic regression, was verified by random assignment into training and test sets and shown to reduce the data dimension and avoid collinearity in CEUS variables. This prospective study consecutively collected 238 patients with 238 4A masses confirmed pathologically. All enrolled patients accepted CEUS examination. The diagnostic performance of senior and junior radiologists, PCR-CEUS and combined methods was compared. The PCR-CEUS system had consistent diagnostic performance in both the training and test sets, with an area under the curve (AUC) of 0.831 (0.765-0.897), 0.798 (0.7034-0.892) and 0.854 (0.765-0.943) (all P > 0.05). The AUC of the combined diagnostic model (PCR-CEUS + Senior radiologists) was higher than that of senior radiologists, and the combined model had higher sensitivity (0.875 (0.781-0.969) vs. 0.729 (0.603-0.855)) without compromising specificity. Furthermore, the AUC and specificity of the combined model (PCR-CEUS + Junior radiologists) (0.852 (0.787-0.916)) was higher than that of junior radiologists (0.665 (0.592-0.737) (P < 0.00001)). PCR-CEUS demonstrated good ability in differentiating malignant BI-RADS-US 4A masses and was helpful for both senior and junior radiologists.     

环境照度对用显示器视读CR影像信息的影响

目的：测试环境照度对用显示器视读CR影像信息的影响。方法：请6名有经验的医学影像诊断医师，在不同环境照度下用显示器视读标准体模TRG的CR影像，获得制作ROC曲线的原始数据，以比较影像信息量多少的ROC曲线面积值Az。结果：在环境照度为100Lx、显示器亮度为60cd/m^2时视读的CR影像上获得ROC曲线面积Az=0.8661，分别比环境照度为12Lx、40Lx、180Lx时的多8%、3%和11%；在环境照度为100Lx、显示器亮度为100cd/m^2时视读的CR影像上获得的ROC曲线面积Az=0.8250，分别比环境照度为12Lx、40Lx、180Lx时的多13%、1%和8%。结论：在同一显示器亮度下视读，环境照度为100Lx时获得的ROC曲线面积Az值最大。     

高分辨率T2WI联合合成双反转恢复图像判断直肠癌T分期北大核心CSCD

目的观察合成MRI(SyMRI)定量参数及高分辨率T2WI(HR-T2WI)联合合成双反转恢复(SyDIR)图用于直肠癌T分期的价值。方法回顾性分析79例经病理证实直肠腺癌,将病理T1及T2期归为肌层未突破组、T3及T4期归为肌层突破组。由2名影像科医师(分别具有低及高年资)基于HR-T2WI及联合SyDIR图行直肠癌T分期,观察其与病理结果的一致性,计算T分期诊断准确率;比较组间SyMRI定量参数值T1、T2及质子密度(PD)的差异。绘制受试者工作特征曲线,计算曲线下面积(AUC),评估HR-T2WI及其联合SyDIR图和T2值诊断直肠癌突破肌层的效能。结果79例中,病理T分期T1期3例,T2期32例,T3期27例,T4期17例。2名医师基于HR-T2WI判断直肠癌T分期与病理结果的一致性均较强(Kappa=0.626、0.784,P均<0.001),诊断准确率分别为75.95%(60/79)及86.08%(68/79);基于HR-T2WI联合SyDIR图判断直肠癌T分期与病理结果的一致性为较强及强(Kappa=0.783、0.829,P均<0.001),诊断准确率分别为84.81%(67/79)及88.61%(70/79)。低年资医师基于HR-T2WI联合SyDIR图判断直肠癌突破肌层的AUC(0.940)高于基于单一HR-T2WI(AUC=0.843,Z=2.443,P=0.015);高年资医师基于HR-T2WI联合SyDIR图的AUC(0.954)与基于单一HR-T2WI差异无统计学意义(AUC=0.943,Z=0.366,P=0.720),后者与低年资医师基于HR-T2WI联合SyDIR图差异无统计学意义(Z=0.108,P=0.918)。2名医师所测肌层未突破组T2值均高于肌层突破组(P均<0.001),其余定量参数差异均无统计学意义(P均>0.05);根据T2值判断直肠癌突破肌层的AUC为0.812。结论SyMRI定量参数T2值对判断直肠癌T分期具有一定价值;HR-T2WI联合SyDIR图有助于提高低年资影像科医师判断直肠癌T分期的效能。     

基于深度学习的乳腺X线摄影钙化检测系统对乳腺可疑钙化的检出效能

目的 评价基于深度学习（DL）的乳腺X线摄影钙化检测系统对乳腺可疑钙化的检出效能。方法 回顾分析932例接受乳腺X线检查患者的头足位（CC）和内外斜位（MLO）资料,由2名低年资医师和DL系统盲法独立阅片,一名高年资医师审核结果。比较DL系统与低年资医师检出敏感度差异,结合双向表χ²检验,评价不同BI-RADS分类、钙化形态和分布影响。结果 针对3 728幅影像（932例）,标记可疑钙化274例。2名低年资医师和DL系统的检出敏感度分别是76.64%（210/274）、82.12%（225/274）和99.64%（273/274）。DL系统检出敏感度不受钙化形态、分布、BI-RADS分类等因素的影响（P均>0.05）,而低年资医师对无定形钙化和团簇分布钙化的敏感度明显降低（P均<0.05）。结论 基于DL的乳腺X线影像钙化检出系统检出对可疑形态钙化的敏感度高且稳定,可辅助临床医师减少漏检。     

Computer-aided diagnosis based on speckle patterns in ultrasound images

For breast ultrasound, the scatterer number density from backscattered echo was demonstrated in previous research to be a useful feature for tumor characterization. To take advantage of the scatterer number density in B-mode images, spatial compound imaging was obtained, and the statistical properties of speckle patterns were analyzed in this study for use in distinguishing between benign and malignant lesions. A total of 137 breast masses (95 benign cases and 42 malignant cases) were used in the proposed computer-aided diagnosis (CAD) system. For each mass, the average number of speckle pixels in a region of interest (ROI) was calculated to use the concept of scatterer number density. In addition, the first-order and second-order statistics of the speckle pixels were quantified to obtain the distributions of the pixel values and the spatial relations among the pixels. The performance of the speckle features extracted from each ROI was compared with the performance of the segmentation features extracted from each segmented tumor. As a result, the proposed CAD system using the speckle features achieved an accuracy of 89.1% (122/137); a sensitivity of 81.0% (34/42); and a specificity of 92.6% (88/95). All of the differences between the speckle features and the segmentation features are not statistically significant (p > 0.05). In a receiver operating characteristic (ROC) curve analysis, the Az value, area under ROC curve, of the speckle features was significantly better than the Az value of the segmentation features (0.93 vs. 0.86, p = 0.0359). The performance of this approach supports the notion that the speckle patterns induced by the scatterers in tissues can provide information for classifying tumors. The proposed speckle features, which were extracted readily from drawing an ROI without any preprocessing, also provide a more efficient classification approach than tumor segmentation.