Differentiating autoimmune pancreatitis from pancreatic ductal adenocarcinoma with CT radiomics features

PurposeThe purpose of this study was to determine whether computed tomography (CT)-based machine learning of radiomics features could help distinguish autoimmune pancreatitis (AIP) from pancreatic ductal adenocarcinoma (PDAC).Materials and MethodsEighty-nine patients with AIP (65 men, 24 women; mean age, 59.7 ± 13.9 [SD] years; range: 21–83 years) and 93 patients with PDAC (68 men, 25 women; mean age, 60.1 ± 12.3 [SD] years; range: 36–86 years) were retrospectively included. All patients had dedicated dual-phase pancreatic protocol CT between 2004 and 2018. Thin-slice images (0.75/0.5 mm thickness/increment) were compared with thick-slices images (3 or 5 mm thickness/increment). Pancreatic regions involved by PDAC or AIP (areas of enlargement, altered enhancement, effacement of pancreatic duct) as well as uninvolved parenchyma were segmented as three-dimensional volumes. Four hundred and thirty-one radiomics features were extracted and a random forest was used to distinguish AIP from PDAC. CT data of 60 AIP and 60 PDAC patients were used for training and those of 29 AIP and 33 PDAC independent patients were used for testing.ResultsThe pancreas was diffusely involved in 37 (37/89; 41.6%) patients with AIP and not diffusely in 52 (52/89; 58.4%) patients. Using machine learning, 95.2% (59/62; 95% confidence interval [CI]: 89.8–100%), 83.9% (52:67; 95% CI: 74.7–93.0%) and 77.4% (48/62; 95% CI: 67.0–87.8%) of the 62 test patients were correctly classified as either having PDAC or AIP with thin-slice venous phase, thin-slice arterial phase, and thick-slice venous phase CT, respectively. Three of the 29 patients with AIP (3/29; 10.3%) were incorrectly classified as having PDAC but all 33 patients with PDAC (33/33; 100%) were correctly classified with thin-slice venous phase with 89.7% sensitivity (26/29; 95% CI: 78.6–100%) and 100% specificity (33/33; 95% CI: 93–100%) for the diagnosis of AIP, 95.2% accuracy (59/62; 95% CI: 89.8–100%) and area under the curve of 0.975 (95% CI: 0.936–1.0).ConclusionsRadiomic features help differentiate AIP from PDAC with an overall accuracy of 95.2%.     

How accurate is atherosclerosis imaging by coronary computed tomography angiography?

Invasive coronary plaque imaging such as intravascular ultrasound and optical coherence tomography has been widely used to observe culprit or non-culprit coronary atherosclerosis, as well as optimize stent sizing, apposition and deployment. Coronary computed tomographic angiography (CTA) is non-invasively available to assess coronary artery disease (CAD) and has become an appropriate strategy to evaluate patients with suspected CAD. Given recent technologies, semi-automated plaque software is available to identify coronary plaque stenosis, volume and characteristics and potentially allows to be used for the assessment of more details of plaque information, progression and future risk as a surrogate tool of the invasive imaging modalities. This review article aims to focus on various evidence in coronary plaque imaging by coronary CTA and describes how accurate coronary CTA can classify coronary atherosclerosis.     

A case of unilateral opercular syndrome associated with a subcortical lesion

Summary A patient who developed a unilateral opercular syndrome following a cerebrovascular accident is described. Computed tomography showed that the lesion did not affect the opercular cortex, but involved deep white matter and the head of the caudate nucleus of the left hemisphere. Persistent hypophonia and transient aphasia were associated. Comparison with previous cases is discussed.     

Focal vulnerability of developing brain: CT studies on disproportionally dilated lateral ventricles

Summary Cranial computed tomography (CT) of 108 cases with dilated lateral ventricles was reviewed to elucidate the relationship between focal vulnerability of developing brain and disproportional dilatation of lateral ventricles. CT findings of 108 cases with symmetrical dilatation of lateral ventricles were classified into three types by morphometry of lateral ventricles: anterior horn predominant type (31 cases), diffuse type (36 cases), posterior horn predominant type (41 cases). Posterior horn predominant type has a tendency to occur in congenital anomalies and premature brain damage, and anterior horn predominant type in infantile brain damage. This disproportional dilatation of anterior or posterior horns suggests a vulnerability of periventricular structure in developing brain.     

Ridge number density: A new parameter for in vivo bone structure analysis

An advanced analysis of the mechanical properties of bone should include information about the microarchitecture of cancellous bone in addition to its density. It has recently been shown that high-resolution quantitative computed tomography and magnetic resonance imaging have the potential to assess such information in a noninvasive way in patients. Both techniques, however, lack sufficient spatial resolution to image the individual trabeculae with true precision. In this work, a new parameter, Ridge number density (RND), is introduced. RND is a measure for the trabecular number, which can be extracted directly from high-resolution three-dimensional (3D) images of patients. We applied the RND technique to a test group of nine healthy, postmenopausal women measured repetitively with a high-resolution 3D peripheral quantitative computed tomography (3D-pQCT) system with 165 × 165 × 165 μm³ voxel size. Simultaneously with the RND determination, the trabecular bone density (TBD) was also assessed in the same volume of interest. The examination site was the distal radius. The intersubject variability of the measured test group was 10.5% for RND and 26.3% for TBD. The root mean square error between first and second examinations (midterm reproducibility) was 1.6% and 1.1%, respectively. RND is determined independently from TBD and pertains to the structure of the cancellous bone. As such, it might add crucial information in cases where bone mass or bone density measurements alone give ambigous results.     

原发性小肠恶性肿瘤的CT诊断

本文回顾性分析３４例经病理证实的原发性小肠恶性肿瘤的ＣＴ表现。病变包括腺癌１６例，平滑肌肉瘤１２例，淋巴瘤６例。ＣＴ扫描能明确肿瘤向腔内和腔外侵犯的程度，清晰显示肿瘤与邻近结构的关系，有无局部及远处转移等，为临床治疗提供帮助。鉴别诊断应包括小肠良性肿瘤、炎性病变和转移性肿瘤等。     

薄层动态CT对胰腺癌的早期诊断

笔者采用薄层动态ＣＴ扫描，诊断直径≤３ｃｍ的胰腺癌１２例，其直接征象为胰头轻度不规则增大或有小的局部隆起、钩突圆隆变形、胰腺实质的分叶形态消失；间接征象为胰管、胆管的梗阻性扩张。认为薄层动态ＣＴ扫描是早期胰腺癌诊断的有效方法。     

Imaging of thoracic and lumbar spinal extradural arachnoid cysts: report of two cases

Extradural arachnoid cysts are uncommon expanding lesions in the spinal canal which may communicate with the subarachnoid space. Usually in the lower thoracic spine, they may cause symptoms by compressing the spinal cord or nerve roots. We report cases of thoracic and lumbar arachnoid cysts studied by cystography, myelography, CT and MRI. These techniques showed extradural cystic lesions containing cerebrospinal fluid, with variable communication with the subarachnoid space, causing anterior displacement and flattening of the spinal cord. Received: 3 November 1995 Accepted: 16 April 1996     

CT and MRI of the middle ear

Summary High-resolution computed tomography (HRCT) provides excellent contrast between osseous structures, air and soft tissue in conjunction with high spatial resolution. Therefore, thin-section HRCT with bone window setting is the method of choice for the examination of the middle ear structures. The indications are acute and chronic inflammatory changes, cholesteatoma and tumor, the “postoperative middle ear”, and malformations. In most cases, HRCT enables differentiation between inflammatory changes, cholesteatoma, and tumor. The excellent depiction of subtle osseous details enables the identification of erosions of the ossicles or of the bony walls of the mastoid cells, of osseous defects of the tegmen, of the bony labyrinth, and of the tympanic course of the facial canal. In addition, HRCT enables excellent depiction of reconstructions of the ossicles or prosthesis of the ossicles. Although HRCT is the first method of choice, magnetic resonance imaging (MRI) may provide additional information and lead to a more accurate diagnosis in some cases. This is explained by the excellent soft tissue contrast provided by MRI. In addition, MRI offers the possibility of using various pulse sequences and the administration of IV contrast material. Therefore, MRI may allow the differentiation between inflammatory changes, cholesteatoma, and tumor in those cases in which accurate diagnosis cannot be made by HRCT. The differentiation between a meningocele or meningoencephalocele and other entities such as tumors or cholesteatoma can be established by MRI. Furthermore, MRI can accurately depict cases of labyrinthitis or of neuritis of the facial nerve or of intracranial disease caused by middle ear processes, while this is not always possible by HRCT. In summary, HRCT of the middle ear is the method of choice, but MRI may provide supplementary information in those cases in which accurate diagnosis cannot be established by HRCT.