The constructed CNN system for detecting gastric cancer could process numerous stored endoscopic images in a very short time with a clinically relevant diagnostic ability. It may be well applicable to daily clinical practice to reduce the burden of endoscopists.

Background and aimsThe role of artificial intelligence in the diagnosis of Helicobacter pylori gastritis based on endoscopic images has not been evaluated. We constructed a convolutional neural network (CNN), and evaluated its ability to diagnose H. pylori infection.MethodsA 22-layer, deep CNN was pre-trained and fine-tuned on a dataset of 32,208 images either positive or negative for H. pylori (first CNN). Another CNN was trained using images classified according to 8 anatomical locations (secondary CNN). A separate test data set (11,481 images from 397 patients) was evaluated by the CNN, and 23 endoscopists, independently.ResultsThe sensitivity, specificity, accuracy, and diagnostic time were 81.9%, 83.4%, 83.1%, and 198 s, respectively, for the first CNN, and 88.9%, 87.4%, 87.7%, and 194 s, respectively, for the secondary CNN. These values for the 23 endoscopists were 79.0%, 83.2%, 82.4%, and 230 ± 65 min (85.2%, 89.3%, 88.6%, and 253 ± 92 min by 6 board-certified endoscopists), respectively. The secondary CNN had a significantly higher accuracy than endoscopists (by 5.3%; 95% CI, 0.3–10.2).ConclusionH. pylori gastritis could be diagnosed based on endoscopic images using CNN with higher accuracy and in a considerably shorter time compared to manual diagnosis by endoscopists.

We found a higher proportion of reports of IP for nivolumab in combination with EGFR-TKI vs treatment with either drug alone. Owing to the limitations of this study, the results warrant further confirmation. However, careful consideration should be given to the possibility of an increased risk of IP when EGFR-TKI is administered in combination with nivolumab, including concomitant and sequential use, and careful monitoring for IP is recommended.

No abstract

Objective Adherence to rehabilitation exercise is much lower in patients with hematologic malignancies (22.5-45.8%) than in patients with solid tumors (60-85%) due to the administration of more intensive chemotherapeutic regimens in the former. Virtual reality exercise can be performed even in a biological clean room and it may improve the adherence rates in elderly patients with hematologic malignancies. Thus, in this pilot study, we aimed to investigate the feasibility and safety of virtual reality exercise intervention using Nintendo Wii Fit in patients with hematologic malignancies receiving chemotherapy. Methods In this feasibility study, 16 hospitalized patients with hematologic malignancies aged ! 60 years performed virtual reality exercise for 20 minutes using the Nintendo Wii Fit once a day, five times a week, from the start of chemotherapy until hospital discharge. The adherence rate, safety, and physical and psychological performances were assessed. Results The adherence rate for all 16 patients was 66.5%. Nine patients completed the virtual reality exercise intervention with 88 sessions, and the adherence rate was 62.0%. No intervention-related adverse effects >Grade 2, according to National Cancer Institute Common Terminology Criteria for Adverse Events version 3.0, were observed. We noted maintenance of the physical performance (e.g., Barthel index, handgrip strength, knee extension strength, one-leg standing time, and the scores of timed up and go test and Instrumental Activities of Daily Living) and psychosocial performance (e.g., score of hospital anxiety and depression scale). Conclusion Virtual reality exercise using the Wii Fit may be feasible, safe and efficacious, as demonstrated in our preliminary results, for patients with hematologic malignancies receiving chemotherapy.

Clinical Practice Guidelines (CPGs) play significant roles in most medical fields. However, little is known about the extent of financial Conflicts of Interest (FCOIs) related to pharmaceutical companies (Pharma) selling dermatology prescription products and dermatology CPG authors in Japan. The aims of this study were to elucidate the characteristics and distribution of payments from Pharma to dermatology CPG authors in Japan, and to evaluate the extent of transparency and accuracy in their FCOI disclosures. We analyzed the records of 296 authors from 32 dermatology CPGs published by the Japanese Dermatological Association from the beginning of 2015 to the end of 2018. Using the payment data reported by 79 Pharma between 2016-2017 in Japan, we investigated the characteristics of the CPG authors and the payments from the Pharma to them. Furthermore, we evaluated the transparency and accuracy of the FCOI disclosures of the individual CPG authors. Of the 296

ObjectiveThis study investigated payments made by pharmaceutical companies to oncology specialists in Japan, what the payments were for and whether the receipt of such payments contravened any conflict of interest (COI) regulations.Design, setting and participantsPayment data to physicians, as reported by all pharmaceutical companies belonging to the Japan Pharmaceutical Manufacturers Association, were retrospectively extracted for 2016. Of the named individual recipients of payments, all certified oncologists were identified, using certification data from the Japanese Society of Medical Oncology (JSMO). The individual specialisations of each of the oncologists was also identified.OutcomePayments to individual cancer specialists and what they were for were identified. Factors associated with receipt of higher value payments and payment flows to specialties were determined. Companies selling oncology drugs with annual sales of ≥5 billion yen (£33.9 million, €40.2 million and $46.0 million) (high revenue-generating drugs) were identified.ResultsIn total, 59 companies made at least one payment to oncologists. Of the 1080 oncology specialists identified, 763 (70.6%) received at least one payment, while 317 received no payment. Of the 763, some 142 (13.1%) receiving at least 1 million yen (£6,800, €8,000 and $9200) accounted for 71.5% of the total. After adjustment of covariates, working for university hospitals and cancer hospitals and male gender were key factors associated with larger monetary payments. Payments preferentially targeted on cancer specialties using high revenue-generating drugs. The JSMO has its own COI policy for its members, but the policy did not mention any specific guidelines for certified oncology specialists.ConclusionFinancial relationships were identified and quantified between pharmaceutical companies and oncology specialists, but the extent and worth varied significantly. Given the frequency and amounts of money involved in such linkages, it would be beneficial for specific COI regulations to be developed and policed for oncologists.

BackgroundLittle information is available concerning how patient delay may be affected by mass disasters. The main objectives of the present study are to identify whether there was a post-disaster increase in the risk of experiencing patient delay among breast cancer patients in an area affected by the 2011 triple disaster in Fukushima, Japan, and to elucidate factors associated with post-disaster patient delay. Sociodemographic factors (age, employment status, cohabitant status and evacuation status), health characteristics, and health access- and disaster-related factors were specifically considered.MethodsRecords of symptomatic breast cancer patients diagnosed from 2005 to 2016 were retrospectively reviewed to calculate risk ratios (RRs) for patient delay in every year post-disaster compared with the pre-disaster baseline. Total and excessive patient delays were respectively defined as three months or more and twelve months or more from symptom recognition to first medical consultation. Logistic regression analysis was conducted for pre- and post-disaster patient delay in order to reveal any factors potentially associated with patient delay, and changes after the disaster.ResultsTwo hundred nineteen breast cancer patients (122 pre-disaster and 97 post-disaster) were included. After adjustments for age, significant post-disaster increases in RRs of experiencing both total (RR: 1.66, 95% Confidence Interval (CI): 1.02–2.70, p < 0.05) and excessive patient delay (RR: 4.49, 95% CI: 1.73–11.65, p < 0.01) were observed. The RRs for total patient delay peaked in the fourth year post-disaster, and significant increases in the risk of excessive patient delay were observed in the second, fourth, and fifth years post-disaster, with more than five times the risk observed pre-disaster. A family history of any cancer was the only factor significantly associated with total patient delay post-disaster (odds ratio: 0.38, 95% CI: 0.15–0.95, p < 0.05), while there were no variables associated with delay pre-disaster.ConclusionsThe triple disaster in Fukushima appears to have led to an increased risk of patient delay among breast cancer patients, and this trend has continued for five years following the disaster.Electronic supplementary materialThe online version of this article (doi:10.1186/s12885-017-3412-4) contains supplementary material, which is available to authorized users.