Anisotropy of Trabecular Bone from Ultra-Distal Radius Digital X-Ray Imaging: Effects on Bone Mineral Density and Age

Background: When applied to trabecular bone X-ray images, the anisotropic properties of trabeculae located at ultra-distal radius were investigated by using the trabecular bone scores (TBS) calculated along directions parallel and perpendicular to the forearm. Methodology: Data from more than two hundred subjects were studied retrospectively. A DXA (GE Lunar Prodigy) scan of the forearm was performed on each subject to measure the bone mineral density (BMD) value at the location of ultra-distal radius, and an X-ray digital image of the same forearm was taken on the same day. The values of trabecular bone score along the direction perpendicular to the forearm, TBSx, and along the direction parallel to the forearm, TBSy, were calculated respectively. The statistics of TBSx and TBSy were calculated, and the anisotropy of the trabecular bone, which was defined as the ratio of TBSy to TBSx and changed with subjects’ BMD and age, was reported and analyzed. Results: The results show that the correlation coefficient between TBSx and TBSy was 0.72 (p BMD and age was reported. The results showed that decreased trabecular bone anisotropy was associated with deceased BMD and increased age in the subject group. Conclusions: This study shows that decreased trabecular bone anisotropy was associated with decreased BMD and increased age.

Full-field mapping of internal strain distribution in red sandstone specimen under compression using digital volumetric speckle photography and X-ray computed tomography

It is always desirable to know the interior deformation pattern when a rock is subjected to mechanicalload. Few experimental techniques exist that can represent full-field three-dimensional （3D） straindistribution inside a rock specimen. And yet it is crucial that this information is available for fully understandingthe failure mechanism of rocks or other geomaterials. In this study, by using the newlydeveloped digital volumetric speckle photography （DVSP） technique in conjunction with X-ray computedtomography （CT） and taking advantage of natural 3D speckles formed inside the rock due to materialimpurities and voids, we can probe the interior of a rock to map its deformation pattern under load andshed light on its failure mechanism. We apply this technique to the analysis of a red sandstone specimenunder increasing uniaxial compressive load applied incrementally. The full-field 3D displacement fieldsare obtained in the specimen as a function of the load, from which both the volumetric and the deviatoricstrain fields are calculated. Strain localization zones which lead to the eventual failure of the rock areidentified. The results indicate that both shear and tension are contributing factors to the failuremechanism. 2015 Institute of Rock and Soil Mechanics, Chinese Academy of Sciences. Production and hosting byElsevier B.V. All rights reserved.

Comparison between Visualization of Microcalcifications by Digital Breast Tomosynthesis and Full-Field Digital Mammography

Objective: To date, few studies have compared the diagnostic performance and visibility of microcalcifications obtained using digital breast tomosynthesis (DBT) with those obtained from full-field digital mammography (FFDM). The visualization and characterization of microcalcifications with DBT remain controversial. The purpose of this study was to compare the visibility of microcalcifications and determine whether DBT exhibits a diagnostic advantage for visualizing microcalcifications over FFDM. Methods: We retrospectively reviewed 120 cases including DBT and FFDM imaging (60 histologically verified as breast cancers and 60 as benign microcalcifications or normal). DBT images with a wide scan-angle of 50° and FFDM images were obtained using a flat-panel system (MAMMOMAT Inspiration, Siemens). Images were independently reviewed by four board-certified radiologists and evaluated for the presence of microcalcifications, probability of malignancy (BI-RADS classification), and visibility. Results: In predicting the malignancy of detected microcalcifications, no significant difference was found between readers’ areas under the receiver operating characteristic curve for DBT and FFDM (p = 0.068). The visibility scores of detected microcalcifications were 3.74 ± 1.06 for DBT and 3.46 ± 0.93 for FFDM, respectively. The visibility of microcalcifications when using DBT was found to be significantly superior to that of FFDM (p < 0.05). Conclusion: Our results suggest that the image quality of DBT with a wide scan-angle is comparable or superior to that obtained with FFDM in terms of both visibility and assessment of microcalcifications.

Rapid digitalization and panoramic evaluation of weld X-ray film

The computer evaluation of weld X-ray film is an attractive technique for weld seam NDT （ nondestructive testing）. To achieve this target, digitalization of film is the first step and automatic defect identification is another key technique. In this paper, a weld X-ray film digitalizing system has been established with linear array CCD and highlight LED light source. Its space resolution can reach 0. 04 mm/pixel and scanning speed can reach 100 mm/s for an industrial film. The transfer function curves of the system have been measured and the results indicate that its image gray resolution can reach 88 G/D at 4. 5D, and its dynamic range can be wider than 2. OD. In order to facilitate the evaluation of large welded structure, a panoramic evaluation algorithm is developed also. The algorithm includes image matching, image fusion and panoramic evaluation of the long linked film image.

Combined use of fine needle aspiration cytology and full field digital mammography in preoperative assessment of breast masses

Objective: To evaluate the accuracy of these two methods and focus on the analysis and management of the false-negative cases. Methods: Results of full field digital mammography (FFDM) and fine needle aspiration cytology (FNAC) were obtained and analyzed from a consecutive of 102 women with palpable breast masses, results were correlated with the histopathological findings. Results: Of the 102 cases, malignancy was confirmed in 43 cases (42.16%) by final pathological examination, the sensitivity and specificity of cancer detection with FNA cytology was 90.7% (39/43) and 89.8% (53/59), re- spectively, the whole accuracy was 90.2% (92/102), with a positive predictive value of 86.7% (39/45) and a negative predictive value of 93.0% (53/57). FFDM gave a sensitivity of 88.4% (38/43), specificity of 83.1% (49/59), and whole accuracy 85.3% (87/102), the positive predictive value and negative predictive value was 79.2% (38/48) and 90.7% (49/54), respectively. All the FNAC-negative cancer cases were suggestive of malignancy by FFDM findings, however, the benign cases which present as equivocal finding by FNA cytology, could not be ruled out the presence of malignancy. Conclusion: FNAC and FFDM both are accurate, effective and economical diagnostic modalities, combined use of these two methods can reduced the misdiag- nosis rate of breast masses.

Relationship between Full-Field Digital Mammographic Features and Clinicopathologic Characteristics in 176 Cases with Breast Cancer

OBJECTIVE Different mammographic features are probably predictive of different prognosis. However, ambiguity still exists in understanding the relationship between them. In resent years, digital mammography has been available for clinical use which has led to a revolution in the resolving of images and an increase in early-stage breast cancer detection. Based on the above knowledge, this study was performed to evaluate the relationship between full-field digital mammographic features and clinicopathologic characteristics in breast cancer. METHODS Digital mammograms of 176 patients with pathologically proven breast cancer were reviewed. Also, clinical and pathologic records （histological types and axillary lymph nodes status） were retrospectively examined. RESULTS Most of the patients with a solitary microcalcification were young women under the age of 50（84.4%）, but the majority of the patients with microcalcifications complicated by a mass were elderly women. Microcalcifications detected by mammography occurred frequently in ductal carcinoma in situ （28.1%） and in early invasive carcinoma （15.6%）. Breast cancers with expression of microcalcifications combined with a spiculate mass had a high metastatic rate of axillary lymph nodes （69.4%）. A high metastatic rate of axillary lymph nodes was also found in the patients with solitary worm-like microcalcifications （57.1%）, solitary spiculate mass （53.7%） and solitary non-worm-like microcalcifications （44.4%）. Simple worm-like microcalcifications accompanied with metastasis of 4 to 9 axillary lymph nodes occurred in 42.9% of the（6/14） cases. The patients with microcalcifications combined by a spiculate mass and with metastasis of 4 to 9 axillary lymph nodes accounted for 27.8% （10/36） of the cases, and those with metastases of 10 and over accounted for 16.7% （6/36）. CONCLUSION Solitary microcalcifications occur frequently in young women and are usually associated with early breast cancer. There is a close relationship between worm-like microcalcifications, a spiculate mass and positive metastases of axillary lymph nodes, which are an index of poor prognosis.

Long Term Reproducibility Check of a Full Field Digital Mammography System

In this study different types of customized phantoms were used to examine a long term reproducibility check of a full field digital mammography system. These phantoms can be distinguished into three types： a PMMA （polimethylmethacrylate） object for an easy signal-to-noise ratio measurement in a defined region, a phantom with particular details of various dimensions （TORMAS） and a phantom for a contrast-detail analysis （CDMAM）. In our center the ARTINIS ANALYZER software has been used since 2011 to weekly perform a CDMAM phantom analysis. This soitware is based on the original CDCOM algorithm suggested by the EUREF, the European Reference Organization for assured quality breast screening and diagnostic services. However the CDMAM phantom is a contrast-detail object designed for image quality type tests and not for long term reproducibility control. Since 2013, as suggested by the guidelines of Veneto region, a TORMAS weekly acquisition and analysis of data has also been performed. The purpose of this study is to discuss the data obtained by using the two different detail phantoms, CDMAM and TORMAS. These results can provide some correlations among the analyzed parameters comparing them with the ones emerging from the PMMA phantom analysis. The study reveals that the basic PMMA phantom is the object that provides the necessary information about the stability of the exposure parameters and the reproducibility of a mammographic equipment in terms of image quality indexes.

Application of X-ray digital tomosynthesis in knee joint trauma examination

Objective: To explore the clinical value of X-ray digital tomosynthesis(DTS)in the diagnosis of knee joint fractures. Methods: A total of 28 cases of thoracic trauma, X-ray film cannot be clearly diagnosed or can confirm the diagnosis but the need for further identification of forensic diagnosis of cases of DTS scan and three-dimensional reconstruction in order to control the study. Results: 1. The reconstructed images after DTS scanning showed that the knee joint fractures were clearly diagnosed, and the detection rate of 28 knee joint fractures was 92.86%. 2. DTS scanning could clearly detect the fresh knee joint fracture's fracture line, corresponding line and broken bones. For the old fracture, DTS scanning could clearly show the condition of the fracture end's healing. Conclusions: The technique is of great value in the diagnosis of knee joint fracture, especially in the examination of complex structure, thick body and review of internal fixation after fracture.

Research on Influence of Electromagnetic Interference on X-Ray Digital Detection

X-ray digital imaging technology has found wide application owing to its advantages of real-time, visualization and rapid imaging. In substations the substantial electromagnetic interference has some influence on the live detection by the X-ray digital imaging technology, hindering the promotion of the technology in the detection of electric equipment. Based on a large number of field tests, the author carded out a series of researches on electromagnetic interference protection measures, image de-noising, and image enhancement algorithms.

Imaging Analysis of Trabecular Bone Texture Based on the Initial Slope of Variogram of Ultra-Distal Radius Digital X-Ray Imaging: Effects on Bone Mineral Density and Age

Background: When applied to trabecular bone X-ray images, a method for analyzing trabecular bone texture based on the initial slope of variogram (ISV) was used to assess the trabecular bone health. Methodology: Data from more than two hundred subjects were retrospectively studied. For each subject, a DXA (GE Lunar Prodigy) scan of the forearm was performed, and bone mineral density (BMD) value was measured at the location of ultra-distal radius, X-ray digital image of the same forearm was taken on the same day, and ISV value over the same location of ultra-distal radius was calculated. Pearson’s correlation coefficients were calculated to examine the correlation between BMD and ISV of the trabecular bones located at the same ultra-distal radius. ISV values changed with subjects’ age were also reported. Results: The results show that ISV value was highly correlated with the DXA-measured BMD of the same trabecular bone located at the ultra-distal radius. The correlation coefficient between ISV and BMD with the 95% confident was 0.79 ± 0.09. They also demonstrated that the age-related changes in trabecular bone health and differentiated age patterns in males and females, respectively. The results showed that the decrease in BMD was accompanied by a decrease in the initial slope of variogram (ISV). Conclusions: This study suggests that ISV might be used to quantitatively evaluate trabecular health for osteoporosis and bone disease diagnosis.

Efficacy of digital breast tomosynthesis combined with magnetic resonance imaging in the diagnosis of early breast cancer

BACKGROUND The incidence and mortality rate of breast cancer in China rank 120th and 163rd,worldwide,respectively.The incidence of breast cancer is on the rise;the risk increases with age but is slightly reduced after menopause.Early screening,diagnosis,and timely determination of the best treatment plan can ensure clinical efficacy and prognosis.AIM To evaluate the clinical value of magnetic resonance imaging(MRI) combined with digital breast tomosynthesis(DBT) in diagnosing early breast cancer and the effect of breast-conserving surgery by arc incision.METHODS This study was divided into two parts.Firstly,110 patients with early breast cancer confirmed by pathological examination and 110 with benign breast diseases diagnosed simultaneously in Changzhi People’s Hospital of Shanxi Province and Shanxi Dayi Hospital from May 2019 to September 2020 were included in the breast cancer group and the benign group,respectively.Both groups underwent DBT and MRI examination,and the pathological results were used as the gold standard to evaluate the effectiveness of the combined application of DBT and MRI in the diagnosis of early breast cancer.Secondly,according to the operation method,110 patients with breast cancer were divided into either a breast-conserving group(69 patients) or a modified radical mastectomy group(41 patients).The surgical effect,cosmetic effect,and quality of life of the two groups were compared.RESULTS Among the 110 cases of breast cancer,66 were of invasive ductal carcinoma(60.00%),and 22 were of ductal carcinoma in situ(20.00%).Among the 110 cases of benign breast tumors,55 were of breast fibromas(50.00%),and 27 were of breast adenosis(24.55%).The sensitivity,specificity,and area under the curve(AUC) of DBT in the differential diagnosis of benign and malignant breast tumors were 73.64%,84.55%,and 0.791,respectively.The sensitivity,specificity,and AUC of MRI in the differential diagnosis of benign and malignant breast tumors were 84.55%,85.45%,and 0.850,respectively.The sensitivity,specificity,and AUC of DBT combined with MRI in the differential diagnosis of benign and malignant breast tumors were 97.27%,93.64%,and 0.955,respectively.The blood loss,operation time and hospitalization time of the breast-conserving group were significantly lower than those of the modified radical treatment group,and the difference was statistically significant(P < 0.05).After 3 mo of observation,the breast cosmetic effect of the breast-conserving group was better than that of the modified radical group,and the difference was statistically significant(P < 0.05).Before surgery,the quality-of-life scores of the breast-conserving and modified radical mastectomy groups did not differ(P > 0.05).Three months after surgery,the quality-of-life scores in both groups were higher than those before surgery(P < 0.05),and the quality-of-life score of the breast-conserving group was higher than that of the modified radical group(P < 0.05).In the observation of tumor recurrence rate two years after the operation,four patients in the breast-conserving group and one in the modified radical treatment group had a postoperative recurrence.There was no significant difference in the recurrence rate between the two groups(χ2 = 0.668,P = 0.414 > 0.05).CONCLUSION MRI combined with DBT in diagnosing early breast cancer can significantly improve the diagnostic efficacy compared with the two alone.Breast-conserving surgery leads to better cosmetic breast effects and reduces the impact of surgery on postoperative quality of life.

Adaptive segmentation of digital mammograms through reinforcement learning

An approach based on reinfocement learning for the automated segmentation is presented. The approach consists of two modules:segmentation module and learning module. The segmentation module uses the region-growing algorithm combined with the smooth filtering and the morphological filtering to segment mammograms. The learning module uses the segmentation output as the feedback to learn to select the optimal parameter settings of the segmentation algorithm according to the image properties using reinforcement learning techniques. The approach can adapt itself to various kinds of mammograms through training and therefore obviates the tedious and error-prone tuning of parameter settings manually. Quantitative test results show that the approach is accurate for several kinds of mammograms. Compared to previously proposed approaches,the approach is more adaptable to different mammograms.

Tracking the phase transformation and microstructural evolution of Sn anode using operando synchrotron X-ray energy-dispersive diffraction and X-ray tomography

Tin(Sn)holds great promise as an anode material for next-generation lithium(Li)ion batteries but suffers from massive volume change and poor cycling performance.To clarify the dynamic chemical and microstructural evolution of Sn anode during lithiation and delithiation,synchrotron X-ray energydispersive diffraction and X-ray tomography are simultaneously employed during Li/Sn cell operation.The intermediate Li-Sn alloy phases during de/lithiation are identified,and their dynamic phase transformation is unraveled which is further correlated with the volume variation of the Sn at particle-and electrode-level.Moreover,we find that the Sn particle expansion/shrinkage induced particle displacement is anisotropic:the displacement perpendicular to the electrode surface(z-axis)is more pronounced compared to the directions(x-and y-axis)along the electrode surface.This anisotropic particle displacement leads to an anisotropic volume variation at the electrode level and eventually generates a net electrode expansion towards the separator after cycling,which could be one of the root causes of mechanical detachment and delamination of electrodes during long-term operation.The unraveled chemical evolution of Li-Sn and deep insights into the microstructural evolution of Sn anode provided here could guide future design and engineering of Sn and other alloy anodes for high energy density Li-and Na-ion batteries.

Head dual energy-computed tomography angiography versus neuro-digital subtraction angiography

Dual-energy X-ray absorptiometry provides two modes of head computed tomography （CT） angiography scanning： neuro-digital subtraction angiography and dual-energy CT angiography （DE-CTA）. Previous studies have compared image quality, radiation exposure, and bone removal between neuro-digital subtraction angiography and DE-CTA. However, the number of cases was relatively small. The present study examined 300 suspected cases of cerebrovascular disease and observed the methods and duration of post-processing, examination time, and data volume. Results demonstrated similar image quality between the two methods, but lower radiation doses and shorter examination time in DE-CTA. DE-CTA allowed for faster and more stable scanning performance and post-processing methods, facilitating accurate and direct diagnosis of cerebrovascular disease.

Integrating digital twins and deep learning for medical image analysis in the era of COVID-19

Background Digital twins are virtual representations of devices and processes that capture the physical properties of the environment and operational algorithms/techniques in the context of medical devices and tech-nologies.Digital twins may allow healthcare organizations to determine methods of improving medical processes,enhancing patient experience,lowering operating expenses,and extending the value of care.During the present COVID-19 pandemic,various medical devices,such as X-rays and CT scan machines and processes,are constantly being used to collect and analyze medical images.When collecting and processing an extensive volume of data in the form of images,machines and processes sometimes suffer from system failures,creating critical issues for hospitals and patients.Methods To address this,we introduce a digital-twin-based smart healthcare system in-tegrated with medical devices to collect information regarding the current health condition,configuration,and maintenance history of the device/machine/system.Furthermore,medical images,that is,X-rays,are analyzed by using a deep-learning model to detect the infection of COVID-19.The designed system is based on the cascade recurrent convolution neural network(RCNN)architecture.In this architecture,the detector stages are deeper and more sequentially selective against small and close false positives.This architecture is a multi-stage extension of the RCNN model and sequentially trained using the output of one stage for training the other.At each stage,the bounding boxes are adjusted to locate a suitable value of the nearest false positives during the training of the different stages.In this manner,the arrangement of detectors is adjusted to increase the intersection over union,overcoming the problem of overfitting.We train the model by using X-ray images as the model was previously trained on another dataset.Results The developed system achieves good accuracy during the detection phase of COVID-19.The experimental outcomes reveal the efficiency of the detection architecture,which yields a mean average precision rate of 0.94.

A cadaveric breast cancer tissue phantom for phase-contrast X-ray imaging applications

Background:As mammography X-ray imaging technologies advance and provide elevated contrast in soft tissues,a need has developed for reliable imaging phantoms for use in system design and component calibration.In advanced imaging modalities such as refraction-based methods,it is critical that developed phantoms capture the biological details seen in clinical precancerous and cancerous cases while minimizing artifacts that may be caused due to phantom production.This work presents the fabrication of a breast tissue imaging phantom from cadaveric breast tissue suitable for use in both transmission and refraction-enhanced imaging systems.Methods:Human cancer cell tumors were grown orthotopically in nude athymic mice and implanted into the fixed tissue while maintaining the native tumor/adipose tissue interface.Results:The resulting human–murine tissue hybrid phantom was mounted on a clear acrylic housing for absorption and refraction X-ray imaging.Digital breast tomosynthesis was also performed.Conclusion:Both attenuation-based imaging and refraction-based imaging of the phantom are presented to confirm the suitability of this phantom's use in both imaging modalities.

基于数字化乳腺X线影像组学预测浸润性乳腺癌腋窝淋巴结转移的多中心研究

目的:探讨基于多中心数字化乳腺X线影像组学预测浸润性乳腺癌腋窝淋巴结转移的临床应用价值。方法:回顾性搜集728例经病理证实的浸润性乳腺癌患者,按照7:3的比例将皖南医学院第一附属医院弋矶山医院413例浸润性乳腺癌患者随机拆分为训练组289例(淋巴结转移阴性197例,淋巴结转移阳性92例)和验证组124例(淋巴结转移阴性85例,淋巴结转移阳性39例),将阜阳市人民医院和太和县人民医院浸润性乳腺癌患者共计315例(淋巴结转移阴性221例,淋巴结转移阳性94例)作为外部测试组。对比分析双乳内外斜位(MLO)和头尾位(CC)图像,选取病变面积较大的数字化乳腺X线图像使用深睿医疗多模态科研平台进行图像分割及影像组学特征提取,采用特征间线性相关性分析与最小绝对收缩和选择算法(LASSO)对组学特征进行降维并使用支持向量机(SVM)分类器构建影像组学预测模型。采用受试者工作特征(ROC)曲线和决策曲线分析(DCA)评价模型性能。结果:最终筛选出8个影像组学特征构建预测模型用于预测浸润性乳腺癌腋窝淋巴结转移,该模型的预测效能在训练组、验证组和外部测试组分别为0.807、0.790和0.753,敏感度、特异度分别为84.8%和61.4%、79.5%和69.4%、44.7%和92.8%。决策曲线证实了该模型的临床实用性。结论:基于数字化乳腺X线影像组学对浸润性乳腺癌腋窝淋巴结转移的预测具有较高效能,对患者制定个性化的治疗方案和预后评估有着重要的临床应用价值。

数字乳腺断层合成摄影和对比增强乳腺X线摄影检查对乳腺BI-RADS 4类病变的诊断价值

目的探讨对比增强乳腺X线摄影(CEM)与全数字化乳腺X线摄影(FFDM)、数字乳腺断层合成摄影(DBT)检查在乳腺BI-RADS 4类病变中大小评估及良恶性鉴别诊断价值。方法选取经病理证实的124例患者(136个BI�RADS 4类病灶),术前均行CEM、FFDM和DBT检查,分别采用Spearman相关分析和Bland-Altman图评价CEM、FFDM和DBT与病理标本测量的病灶最大径的相关性和一致性;采用受试者工作特征曲线(ROC)评价CEM、FFDM和DBT鉴别乳腺BI-RADS 4类良恶性病变的诊断效能。结果CEM、FFDM和DBT与病理测量病灶大小的相关系数分别为r=0.902,r=0.856和r=0.892(均P<0.001)。CEM、FFDM、DBT测量的病灶最大径与手术病理测量值的平均差值分别为0.24 cm、0.21 cm和0.11 cm。CEM、FFDM、DBT对乳腺良恶性病变诊断的敏感度分别为85.51%、65.22%、79.71%,特异度分别为88.06%、77.61%、77.61%,准确度分别为86.76%、71.32%、78.68%。两两比较结果显示,CEM的ROC曲线下面积(area under the curve,AUC)均高于FFDM、DBT,差异有统计学意义(均P<0.01),而DBT与FFDM的AUC差异无统计学意义(P>0.05)。结论DBT对BI-RADS 4类乳腺病变的大小评估较FFDM、CEM更准确,CEM对乳腺BI�RADS 4类良恶性病变的鉴别诊断效能优于FFDM、DBT。

数字乳腺断层摄影评估新辅助化疗后肿瘤大小的价值

目的评估新辅助化疗后数字乳腺断层摄影(DBT)、全数字化乳腺摄影(FFDM)和超声(US)检查预测残留肿瘤大小的能力。方法回顾性分析2020年1月1日至2021年12月31日病理结果确诊为乳腺癌,行新辅助化疗并接受手术治疗患者的临床及影像学资料。所有患者在新辅助化疗后均行乳腺DBT、FFDM及US检查。采用Pearson相关分析计算不同影像检查方法测量残留肿瘤大小与病理测量结果的相关性;以0.5 cm为阈值,分析不同影像检查预测残留肿瘤大小的准确率。结果85例患者,共86个病灶,去除新辅助化疗后无残留癌的病灶,共纳入72个病灶。新辅助化疗后,DBT、FFDM、US测量残留肿瘤大小与病理测量结果均呈中等程度相关。DBT、FFDM、US预测残留肿瘤大小的准确率分别为59.72%(43/72)、43.06%(31/72)、41.67%(30/72)。DBT与FFDM、US预测残留肿瘤大小的准确率比较,差异具有统计学意义(χ^(2)=4.003、4.695,P=0.045、0.030)。结论新辅助化疗后,DBT预测残留肿瘤大小的准确率优于FFDM与US。DBT在新辅助化疗后的术前评估具有良好的应用前景。

基于数字化乳腺X线影像组学列线图预测浸润性乳腺癌组织学分级的多中心研究

目的:探讨基于多中心数字化乳腺X线影像组学列线图预测浸润性乳腺癌组织学分级的价值。方法:以病理诊断为金标准,按照7:3的比例将皖南医学院第一附属医院弋矶山医院437例浸润性乳腺癌患者随机拆分为训练组305例(Ⅰ/Ⅱ级217例,Ⅲ级88例)和验证组132例(Ⅰ/Ⅱ级94例,Ⅲ级38例),将阜阳市人民医院(n=129)和太和县人民医院(n=162)291例浸润性乳腺癌患者(Ⅰ/Ⅱ级203例,Ⅲ级88例)作为外部测试组。对比分析双乳内外斜位(MLO)和头尾位(CC)图像,选取病变面积较大的数字化乳腺X线图像使用深睿医疗多模态科研平台进行图像分割和影像组学特征提取,采用特征间线性相关性分析与最小绝对收缩和选择算法对组学特征进行降维并使用逻辑回归构建影像组学模型。临床指标经单因素及多因素二元Logistic回归分析并构建临床模型。影像组学评分联合临床指标构建列线图。采用受试者工作特征(ROC)曲线和决策曲线分析(DCA)评价模型性能,使用Delong检验比较模型间的预测效能。结果:最终筛选出3个与浸润性乳腺癌组织学分级最相关的影像组学特征。列线图对浸润性乳腺癌组织学分级的预测效能在训练组、验证组和外部测试组分别为0.811、0.825和0.803,诊断效能优于单一模型。DCA显示在概率值为20%~60%时,训练组、验证组及外部测试组列线图预测浸润性乳腺癌组织学分级的净收益高于影像组学模型及临床模型。结论:基于数字乳腺X线影像组学模型对浸润性乳腺癌组织学分级的预测具有较高的效能,对患者制定个性化治疗方案和预后评估有着重要的临床应用价值。