Comparative assessment of noise properties for two deep learning CT image reconstruction techniques and filtered back projection.

BACKGROUND: Two deep learning image reconstruction (DLIR) techniques from two different computed tomography (CT) vendors have recently been introduced into clinical practice. PURPOSE: To characterize the noise properties of two DLIR techniques with different training methods, using a phantom containing a simple uniform and a complex non-uniform region. METHODS: A water-bath phantom with a diameter of 300 mm was used as a base phantom. A textured phantom with a diameter of 128 mm, which was made of two materials, one equivalent to water and the other being 12 mg/ml diluted iodine, irregularly mixed to create a complex texture (non-uniform region), was placed in the base phantom. Thirty repeated phantom scans were performed using two CT scanners (Revolution CT with Apex Edition, GE Healthcare; Aquilion One PRISM Edition, Canon Medical Systems) at two dose levels (CT dose index: 5 and 15 mGy). Images were reconstructed with each CT system's filtered back projection (FBP) and DLIR [TrueFidelity (TF), GE Healthcare; Advanced intelligent Clear-IQ Engine Body Sharp (AC), Canon Medical Systems] for three process strengths. For basic characteristics of noise, the standard deviation (SD) and noise power spectrum (NPS) were measured for the uniform (water) region. A noise magnitude map was generated by calculating the inter-image SD at each pixel position across the 30 images. Then, a noise reduction map (NRM), which visualizes the relative differences in noise magnitude between FBP and DLIR, was calculated. The NRM values ranged from 0.0 to 1.0. A low NRM value represents a less aggressive noise reduction. The histograms of the NRM value were analyzed for the uniform and non-uniform regions. RESULTS: The reduction in noise magnitude compared with FBP tended to be greater with AC (45%-85%) than with TF (32%-65%). The average NPS frequencies of TF and AC were almost comparable to those of FBP, except for the low-dose condition and the high noise reduction strength for AC. The NRM values of TF and AC were higher in the uniform region than in the non-uniform region. In the non-uniform region, TF's average NRM values (0.21-0.48) tended to be lower than AC's (0.39-0.78). The histograms for TF showed a small overlap between the uniform and the non-uniform regions; in contrast, those for AC showed a greater overlap. This difference seems to indicate that TF processes the uniform and non-uniform regions more differently than AC does. CONCLUSION: This study has revealed a distinct difference in characteristics between the two DLIR techniques: TF tends to offer less aggressive noise reduction in non-uniform regions and preserve the original signals, whereas AC tends to prioritize noise filtering over edge-preservation, especially at the low-dose condition and with the high noise reduction strength.

Immunohistochemical staining for IMP3 in patients with duodenal papilla tumors: assessment of the potential for diagnosing endoscopic resectability and predicting prognosis.

BACKGROUND: Endoscopic papillectomy of duodenal papillary tumors (PT) is indicated for adenomas or well-differentiated adenocarcinomas that do not involve the sphincter of Oddi. However, there is currently no reliable pre-operative method to diagnose the infiltration in the sphincter of Oddi.' Insulin-like growth factor 2 mRNA protein 3 (IMP3) staining is reportedly associated with advanced disease stage and clinical outcomes in many carcinomas. The aim of this retrospective study was to investigate the ability of diagnosing sphincter of Oddi involvement in PT and predicting the prognoses using IMP3 immunohistochemistry. METHODS: Twenty-five resected specimens from patients with PT and 24 biopsy specimens from the same patients excluding one were immunostained for IMP3. The percentage of positive cells in the tumor was evaluated and compared with the final pathological diagnosis and prognosis. RESULTS: The final pathological diagnoses were adenoma in 5 patients and adenocarcinoma in 20 patients (no sphincter of Oddi involvement in 5 and involvement in 15). The ability to diagnose sphincter of Oddi involvement based on the percentage of IMP3-positive cells in resected specimens and tissue biopsies was the area under the curve 0.8 and 0.78, respectively, of the receiver operating characteristic curve, and the accuracies were 80.0% and 75.0% (cutoff value: 10%), respectively. Moreover, patients with an IMP3-positive cell rate of ≥ 10% had a significantly worse prognosis (log-rank test P = 0.01). CONCLUSION: IMP3 immunostaining of resected and biopsy specimens from PT patients enables the diagnosis of sphincter of Oddi involvement objectively and is also effective in predicting the prognosis.

Endoscopic Ultrasound Elastography as a Novel Diagnostic Method for the Assessment of Hardness and Depth of Invasion in Colorectal Neoplasms.

INTRODUCTION: We aimed to compare the efficacy of endoscopic ultrasound elastography (EUS-EG) with that of magnifying chromoendoscopy (MCE) and endoscopic ultrasonography (EUS) for the diagnosis of the depth of invasion in colorectal neoplasms. This is an important clinical issue as the depth of invasion is associated with the risk of metastasis. METHODS: Consecutive patients with suspected superficial colorectal neoplasms, evaluated by MCE, EUS, and EUS-EG, for whom endoscopic submucosal dissection was considered, were enrolled in 2018 (derivation study) and in 2019-2020 (validation study). The primary clinical endpoint was the diagnostic yield differentiating intramucosal and shallow submucosal neoplasms from deep submucosal (dSM) and advanced colorectal cancers. In addition, inter- and intra-observer agreements of the elastic score of colorectal neoplasm (ES-CRN) were evaluated by 2 expert and 2 non-expert endoscopists. RESULTS: Thirty-one (33 lesions) and 50 (55 lesions) patients were enrolled in the derivation and validation studies, respectively. Sensitivity, specificity, positive, and negative predictive values, and accuracy of assessment of the depth of submucosal or deeper invasion in the derivation and validation groups were as follows: EUS-EG, 100/88.2/86.7/100/93.3% and 77.8/86.1/73.7/88.6/83.3%; MCE, 66.7/94.4/90.9/77.3/81.8% and 84.2/91.4/84.2/91.4/88.9%; and EUS, 93.3/77.8/77.8/93.3/84.8% and 89.5/65.7/58.6/92.0/74.1%, respectively. For the 2 expert endoscopists, interobserver agreement for the ES-CRN (first and second assessments) in the derivation group was 0.84 and 0.78, respectively; these values were 0.73 and 0.49, respectively, for the 2 non-expert endoscopists. DISCUSSION/CONCLUSION: All 3 modalities presented similar diagnostic yield. Inter- and intra-observer agreements of the ES-CRN were substantial, even for non-expert endoscopists. Therefore, EUS-EG may be a useful modality in determining the depth of invasion in colorectal neoplasms.

Assessment of lateral hindfoot impingement with weightbearing multiplanar imaging in a flatfoot.

BACKGROUND: Estimation of the lateral hindfoot impingement in the standing position in conventional radiography can be difficult due to superimposition of different bones. Patients with flat feet frequently suffer from pain around the lateral malleolus and sinus tarsi caused by osseous impingement in the lateral hindfoot. Weightbearing multiplanar images (tomosynthesis) yield tomographic images and can be taken while full weightbearing. PURPOSE: To assess the availability of tomosynthesis to determine hindfoot lateral impingement. MATERIAL AND METHODS: A total of 14 feet (in 13 patients) with acquired flatfoot deformity and lateral hindfoot pain were included (mean age 64 years; age range 55-80 years). All patients underwent tomosynthesis, radiography, and computed tomography (CT) (non-weightbearing). Talofibular, calcaneofibular, and talocalcaneal impingement were determined. To compare the number of impingements or to determine the area between each image, statistical evaluations were analyzed using the Mann-Whitney U-test (P < 0.05). RESULTS: On tomosynthesis, we clearly found talofibular impingement in three feet, calcaneofibular impingement in seven feet, and talocalcaneal impingement in 11 feet. Therefore, we could identify most impingements as "positive" compared to those on normal radiographs and CT images. The number of impingements in the calcaneofibular and talocalcaneal regions was significantly higher using tomosynthesis than when using CT (P < 0.05). CONCLUSION: Tomosynthesis imaging makes it easier to obtain CT-like images in a short period of time, in a free position, including while standing, and provides useful information to assess lateral pain in patients with flatfoot deformity.

Feasibility and usefulness of endoscopic ultrasonography-guided shear-wave measurement for assessment of autoimmune pancreatitis activity: a prospective exploratory study.

PURPOSE: To assess the feasibility and the clinical usefulness of a newly developed endoscopic ultrasonography (EUS) shear-wave elastography technique (EUS shear-wave measurement: EUS-SWM) in the diagnosis and treatment of autoimmune pancreatitis (AIP). METHODS: Tissue elasticity was measured in the pancreas in 160 patients. The success rate of EUS-SWMs, the velocity of the shear wave (Vs, m/s), and the reliability index of the Vs measurement (VsN) were evaluated, and the elasticity (median Vs) was compared between AIP patients (n = 14) and normal controls. RESULTS: A total of 3837 EUS-SWMs were performed without adverse events. Overall, 97.6% (3743/3837) were successful. The median VsN was 74%. The median Vs values of the pancreas were as follows: 2.22 m/s in the pancreatic head (push position), 2.36 m/s in the head (pull position), 1.99 m/s in the body, and 2.25 m/s in the tail. The median Vs of the AIP group (2.57 m/s) was significantly higher than that of the normal controls (1.89 m/s) (P = 0.0185). The mean Vs significantly decreased from 3.32 m/s to 2.46 m/s after steroid therapy (n = 6) (P = 0.0234). CONCLUSION: EUS-SWM is feasible and generates credible results. EUS-SWM was a useful method for assessment of the effect of steroid therapy in AIP patients.

Clinical Impact of EUS-Guided Fine Needle Biopsy Using a Novel Franseen Needle for Histological Assessment of Pancreatic Diseases.

Background and Aims: Several studies have shown the benefits of endoscopic ultrasound-guided fine needle biopsy (EUS-FNB) using a Franseen needle for histological assessment. However, studies focusing on pancreatic diseases are limited and the safety of this method has not been well assessed. We aimed to assess the current status and issues of EUS-FNB in the diagnosis of pancreatic diseases. Materials and Methods: We retrospectively reviewed 87 consecutive EUS-FNB specimens using either a 22-gauge Franseen needle (Group A, N = 51) or a conventional 22-gauge fine-needle aspiration needle (Group B, N = 36) for pancreatic diseases, and the diagnostic accuracy and safety were compared. Final diagnoses were obtained based on surgical pathology or a minimum six-month clinical follow-up. Results: Although the diagnostic accuracy for malignancy was 96.1% in Group A versus 88.9% in Group B, with no statistically significant difference (P = 0.19), the median sample area was significantly larger in Group A (4.07 versus 1.31mm2, P < 0.0001). There were no differences between the two needles in the locations from which the specimens were obtained. Adverse events occurred in one case (2%) in Group A (mild pancreatitis) and none in Group B with no statistical significance (P = 0.586). Although there was no case of bleeding defined as adverse events, 2 cases in Group A showed active bleeding during the procedure with increase in the echo-free space, which required CT scanning to rule out extravasation. Eventually, the bleeding stopped spontaneously. Conclusions: Given its guaranteed ability to obtain core specimens and comparable safety, and although the risk of bleeding should be kept in mind, EUS-FNB using a Franseen needle is likely to become a standard procedure for obtaining pancreatic tissue in the near future.