Computer-aided diagnosis of prostate cancer on magnetic resonance imaging using a convolutional neural network algorithm.

OBJECTIVE: To develop a computer-aided diagnosis (CAD) algorithm with a deep learning architecture for detecting prostate cancer on magnetic resonance imaging (MRI) to promote global standardisation and diminish variation in the interpretation of prostate MRI. PATIENTS AND METHODS: We retrospectively reviewed data from 335 patients with a prostate-specific antigen level of <20 ng/mL who underwent MRI and extended systematic prostate biopsy with or without MRI-targeted biopsy. The data were divided into a training data set (n = 301), which was used to develop the CAD algorithm, and two evaluation data sets (n = 34). A deep convolutional neural network (CNN) was trained using MR images labelled as 'cancer' or 'no cancer' confirmed by the above-mentioned biopsy. Using the CAD algorithm that showed the best diagnostic accuracy with the two evaluation data sets, the data set not used for evaluation was analysed, and receiver operating curve analysis was performed. RESULTS: Graphics processing unit computing required 5.5 h to learn to analyse 2 million images. The time required for the CAD algorithm to evaluate a new image was 30 ms/image. The two algorithms showed area under the curve values of 0.645 and 0.636, respectively, in the validation data sets. The number of patients mistakenly diagnosed as having cancer was 16/17 patients and seven of 17 patients in the two validation data sets, respectively. Zero and two oversights were found in the two validation data sets, respectively. CONCLUSION: We developed a CAD system using a CNN algorithm for the fully automated detection of prostate cancer using MRI, which has the potential to provide reproducible interpretation and a greater level of standardisation and consistency.

Comparison of fluorescence optical imaging, ultrasonography and clinical examination with magnetic resonance imaging as a reference in active rheumatoid arthritis patients.

BACKGROUND: Fluorescence optical imaging with indocyanine-green enhancement (FOI) is a new imaging modality for the assessment of hand arthritis. The objective of this study was to compare performance profiles of clinical examination (CE), US and FOI using MRI as a reference in the same active rheumatoid arthritis (RA) patients. METHODS: CE, US, FOI and MRI were performed on six subjects with active RA. Each sequence of FOI was divided into three phases based on indocyanine-green dynamics and the joints were graded semi-quantitatively. Sensitivities and specificities of CE, US and FOI were calculated using the RAMRIS synovitis score >0 as a reference in a total of 30 joints (the second to fifth metacarpophalangeal (MCP) joints and the wrist of the clinically dominant hand). RESULTS: FOI showed sensitivities and specificities, respectively, of 85% and of 94% for Phase-1 and 69% and 94% for Phase-2. Sensitivities and specificities were 100% and 35% for CE (tender or swollen), 92% and 41% for gray scale US, and 77% and 100% for color-Doppler US. CONCLUSIONS: The performance characteristics of FOI in detection of synovitis in patients with active RA are comparable to those of US and more specific than CE. FOI has a potential as an assessment modality of RA.

Prediction of histological grade of hepatocellular carcinoma using quantitative diffusion-weighted MRI: a retrospective multivendor study.

OBJECTIVE: To evaluate the usefulness of quantitative diffusion-weighted (DW) imaging acquired by multivendor magnetic resonance units for predicting grade of hepatocellular carcinoma (HCC). METHODS: 83 patients with 100 histologically diagnosed HCCs who underwent pre-operative liver DW imaging with b = 0 and1000 s mm-2 or b = 0 and800 s mm-2 at any of six institutions were included. Two radiologists independently measured the apparent diffusion coefficient (ADC) of the lesion as well as non-ADC parameters, such as the relative contrast ratio and the contrast-to-noise ratio (CNR) between the lesion and the liver parenchyma on high b-value DW images. The diagnostic performance of the DW parameters in discriminating poorly-differentiated HCCs was compared using receiver operating characteristic (ROC) analysis. RESULTS: The areas under the receiver operating characteristic curves for the CNR (86.4% [95% confidence interval (CI) (77.2-95.6] and 83.9% [95% CI 71.2-96.6] for b = 1000 and 800 s mm-2, respectively] and the relative contrast ratio (85.3% [95% CI 75.5-94.8] and 83.5% [95% CI 70.5-96.4]) tended to be superior to the ADC [71.1% [95% CI (56.9-85.2)] and 75.7% [95% CI (55.1-96.2)]; p < 0.05 for CNR vs ADC for b = 1000 s mm-2, but not significant for other parameters) for discrimination of poorly-differentiated HCCs. CONCLUSION: All DW parameters could discriminate HCC grade. Non-ADC parameters might be more useful than the ADC for predicting poorly-differentiated HCCs. Advances in knowledge: The utility of quantitative DW parameters for predicting HCC grade was demonstrated by using multivendor MR units.

Impact of the Prostate Imaging Reporting and Data System, Version 2, on MRI Diagnosis for Extracapsular Extension of Prostate Cancer.

OBJECTIVE: The purpose of this study is to validate the Prostate Imaging Reporting and Data System, version 2 (PI-RADSv2), in assessing extracapsular extension (ECE), compared with PI-RADS, version 1 (PI-RADSv1). MATERIALS AND METHODS: A total of 210 patients with clinically localized prostate cancer underwent MRI and radical prostatectomy. Two readers independently interpreted the MR images. In PI-RADSv1, 5-point ECE risk scoring was used. In PI-RADSv2, ECE criteria included morphologic features and a tumor-capsule contact length (CL) greater than 10 mm. The diagnostic performance of each PI-RADS version and the cutoff CL were evaluated. RESULTS: ECE was found in 56 patients (26.7%). In PI-RADSv1, maximal accuracy was achieved with a risk score of 3 or greater. At this threshold, positive findings on PI-RADSv1 and PI-RADSv2 were identified in 21.0-34.3% and 49.0-51.4% of patients, respectively. Compared with PI-RADSv1, PI-RADSv2 had higher negative predictive values (84.9-89.1% vs 96.3-97.1%, respectively; p = 0.003 and 0.021, for each reader). PI-RADSv1 and PI-RADSv2 had positive predictive values of 56.9-70.5% and 49.1-50.5%, respectively (p = 0.025 and 0.300, respectively). Interobserver kappa values for PI-RADSv1 and PI-RADSv2 were 0.511 and 0.781, respectively. The best cutoff CL was greater than 10 mm among patients without morphologic features of ECE. For patients positive for ECE on the basis of PI-RADSv2 but not PI-RADSv1, 73.3-74.1% of prostate cancer cases with a biopsy Gleason score of 7 or less and 35.7-44.4% of cases with a biopsy Gleason score of 8 or higher were overstaged. CONCLUSION: PI-RADSv2 reduces understaging and improves interobserver agreement in ECE assessment. However, overstaging is a concern, and the biopsy Gleason score may have a complementary role in reducing overstaging.

IgG4-related inflammatory pseudotumors mimicking multiple meningiomas.

IgG4-related disease is an emerging clinicopathologic entity. Hypophysitis, diffuse thickening of dura, and enlargement of the trigeminal nerve are well-known intracranial involvements of IgG4-related disease. This report of a case of systemic IgG4-related disease is the first to present neuroimaging of apparent supratentorial meningioma-like lesions and thickening and contrast enhancement of the walls of the intracranial internal carotid arteries. It is important to recognize IgG4-related intracranial pseudotumors so that patients do not undergo unnecessary surgical procedures.

Correlation between single nucleotide polymorphisms and jejunal crypt cell apoptosis after whole body irradiation.

PURPOSE: To identify loci concerned with radiosensitivity in a mouse model using single nucleotide polymorphism (SNP) markers. MATERIALS AND METHODS: We subjected 276 second filial generation (F2) mice descended from two inbred mouse strains, radiation-induced apoptosis sensitive C57BL/6JNrs (B6) and radiation-induced apoptosis resistant C3H/HeNrs (C3H), to 2.5 Gy whole-body irradiation. We quantified jejunal crypt apoptosis, performed a genome-wide survey, and identified quantitative trait loci (QTL) associated with radiation sensitivity. We expressed apoptosis levels as an apoptotic score (AS), which was equal to the number of apoptotic bodies divided by the number of crypts. We genotyped the mice for 109 SNP markers. RESULTS: AS values were 97.7+/-32.9 in B6 mice and 49.0+/-24.9 in C3H mice (p < 0.01). Genome-wide analysis revealed 8 markers (2 on chromosome 9, 4 on 15, 1 on 17, and 1 on 18) affecting radiation-induced jejunal apoptosis with log odds (LOD) scores ranging from 2.11+/-3.91. We found a significant locus on chromosome 15, which was previously reported by Weil and colleagues. CONCLUSION: These findings support the view that the radiosensitivity of clinically normal tissue depends on variations in several genes.

Flow direction of ascending lumbar veins on magnetic resonance angiography and venography: would "descending lumbar veins" be a more precise name physiologically?

BACKGROUND: Physiological flow direction of ascending lumbar vein (ALV) is not well recognized. METHODS: Two-dimensional time-of-flight magnetic resonance angiography (MRA) examinations of the lower extremities in 44 patients and venography (MRV) in 59 patients were retrospectively reviewed. chi2 analysis was used to compare the frequency of ALV detection between the MRA and MRV groups and between cases with filling defects above the ALV confluence and other cases in the MRV group. RESULTS: Frequency of ALV detection was significantly higher in the MRA group (60 of 88 veins, 68.2%) than in the MRV group (9 of 118 veins, 7.6%, P < 0.0001) and in cases with filling defects above the ALV confluence (8 of 23 veins, 34.8%; 6 were compression of the left common iliac vein by the right common iliac artery, 2 were thrombus of the proximal bilateral common iliac veins) than in other cases (1 of 95 veins, 1.1%) in the MRV group (P < 0.0001). CONCLUSIONS: Without compression or occlusion above the ALV confluence, the general flow direction of the ALVs is not ascending but descending, suggesting that "descending lumbar veins" is a more physiologically precise name for these veins than ALVs.

Defective repair of radiation-induced DNA damage is complemented by a CHORI-230-65K18 BAC clone on rat chromosome 4.

The Long Evans cinnamon (LEC) rat is highly susceptible to X-irradiation due to defective DNA repair and is thus a model for hepatocellular carcinogenesis. We constructed a bacterial artificial chromosome (BAC) contig of rat chromosome 4 completely covering the region associated with radiation susceptibility. We used transient and stable transfections to demonstrate that defective DNA repair in LEC cells is fully complemented by a 200-kb BAC, CHORI-230-65K18. Further analysis showed that the region associated with radiation susceptibility is located in a 128,543-bp region of 65K18 that includes the known gene Rpn1. However, neither knockdown nor overexpression of Rpn1 indicated that this gene is associated with radiation susceptibility. We also mapped three ESTs (TC523872, TC533727, and CB607546) in the 128,543-bp region, suggesting that 65K18 contains an unknown gene associated with X-ray susceptibility in the LEC rat.

Fine mapping of radiation susceptibility and gene expression analysis of LEC congenic rat lines.

LEC rats constitute an animal model of high susceptibility to X-rays. We developed congenic LEC rat lines (recipient strain, Fischer 344 (F344)) and performed genome-wide genotyping to identify radiation susceptibility genes. We mapped seven positional candidate genes, Bmp10, Gpr73, Gp9, Cnbp, Copg, Rab7, and Rpn1, to an approximately 1.2-Mb region located between loci D4Got85 and D4Got148 on chromosome 4. None of the seven genes has been reported to be associated with radiation susceptibility. Comparison of the coding sequences for these seven genes in F344 and LEC rats showed no changes in deduced amino acid sequences. We determined gene expression differences in Gp73, Gp9, and Cnbp as well as strain-specific variations in upstream sequences of these genes. Our results suggest that radiation susceptibility in the LEC rat is primarily attributable to one of the genes within this approximately 1.2-Mb region; however, expression analysis gave no clear indication as to which gene is responsible.

Effective laparoscopic drainage for intra-abdominal abscess not amenable to percutaneous approach: report of two cases.

PURPOSE: The usefulness of the laparoscopic approach for massive intra-abdominal abscesses is controversial. We report two patients who underwent laparoscopic abscess drainage for massive intra-abdominal abscesses not amenable to the percutaneous approach that were suspected to be caused by acute appendicitis. METHODS: In both patients, four ports were placed at their abdominal walls under general anesthesia. Intra-abdominal abscess cavities were punched out, and the purulent exudates that spilled out from the cavities were aspirated completely. Copious irrigation was performed under direct vision. These procedures were completed laparoscopically. RESULTS: The postoperative clinical courses of the patients were uneventful. The intra-abdominal abscesses did not recur, and no wound complications were recognized. The patients were discharged from our hospital in excellent condition within two weeks. CONCLUSIONS: Laparoscopic drainage for massive intra-abdominal abscess is a minimally invasive and useful procedure compared with the open method or the percutaneous approach. It offers the advantage of being able to explore of the abdominal cavity without an unnecessary wide incision, and aspiration of a purulent exudate is possible under direct vision.

[Study of indoleamine 2,3-dioxygenase expression in patients of esophageal squamous cell carcinoma].

We evaluated the clinical significance of indoleamine 2,3-dioxygenase (IDO) in esophageal squamous cell carcinomas. Operative specimens obtained from 30 patients with esophageal squamous cell carcinomas were investigated by semiquantitative RT-PCR with specific primers against IDO. The correlations among IDO expression, clinicopathologic factors and prognosis were studied. The expression of IDO was observed in 100% of both of the cancer specimens and the normal mucosa specimens. The IDO expression of the cancer specimens was higher than the normal mucosa specimens. The expression of IDO did not correlate to histological classification, growth pattern, lymphatic invasion, venous invasion, or lymph nodes metastasis, but correlated to clinicopathological stage, the value of immunosuppressive acidic protein (IAP). The group with higher levels of IDO expression had a worse survival rate than the IDO expression group with lower levels. The serum IDO levels of cancer patients were higher than healthy donors measured by semiquantitative RT-PCR and HPLC. It is suggested that the expression of IDO in esophageal squamous cell carcinoma patients may play a pivotal role for immunosuppression of those patients.

Immunomodulatory effects of low dose cis-Diaminedichloroplatinum (cisplatin) combined with UFT and PSK in patients with advanced colorectal cancer.

It is well known that cell-mediated immunity is suppressed in patients with neoplastic diseases. We have reported that soluble receptors for interleukin-2 (sIL-2R) and tumor necrosis factor (sTNF-R1) are elevated in the serum of patients with advanced colorectal cancer. The presence of these soluble receptors and immunosuppressive cytokines, including interleukin-10 (IL-10), might be important in the mechanisms of immunosuppression. cis-Diaminedichloroplatinum (cisplatin) has been reported to immunomodulate, especially when used in low dose in combination with 5-Fluorouracil (5-FU). In this study, cisplatin and UFT, a form of uracil and tegafur which is a prodrug of 5-FU, were administered with immunomodulator Polysaccharide K (PSK) to ten patients with colorectal cancer, who showed distant metastasis in the liver or lung, and the serum levels of sIL-2R and sTNF-R1 and the production of gamma-interferon (gamma-INF) and interleukin-10 by peripheral blood mononuclear cells were measured. The serum concentrations of sIL-2R and the production of IL-10 were reduced (p < 0.05) after 2 months of treatment. Thus, this combination appeared to have immunomodulative potential in patients with advanced colorectal cancer.