Search

The present study presents an alternative analytical workflow that combines mid‐infrared (MIR) microscopic imaging and deep learning to diagnose human lymphoma and differentiate between small and large cell lymphoma. We could show that using a deep learning approach to analyze MIR hyperspectral data obtained from benign and malignant lymph node pathology results in high accuracy for correct classification, learning the distinct region of 3900 to 850 cm−1. The accuracy is above 95% for every pair of malignant lymphoid tissue and still above 90% for the distinction between benign and malignant lymphoid tissue for binary classification. These results demonstrate that a preliminary diagnosis and subtyping of human lymphoma could be streamlined by applying a deep learning approach to analyze MIR spectroscopic data. [ABSTRACT FROM AUTHOR]

BackgroundPatient portals offer the possibility to assess patient-reported outcome measures (PROMs) remotely, and first evidence has demonstrated their potential benefits. ObjectiveIn this study, we evaluated patient use of a web-based patient portal that provides patient information and allows online completion of PROMs. A particular focus was on patient motivation for (not) using the portal. The portal was developed to supplement routine monitoring at the Department of Internal Medicine V in Innsbruck. MethodsWe included patients with multiple myeloma and chronic lymphocytic leukemia who were already participating in routine monitoring at the hospital for use of the patient portal. Patients were introduced to the portal and asked to complete questionnaires prior to their next hospital visits. We used system access logs and 3 consecutive semistructured interviews to analyze patient use and evaluation of the portal. ResultsBetween July 2017 and August 2020, we approached 122 patients for participation in the study, of whom 83.6% (102/122) consented to use the patient portal. Patients were on average 60 (SD 10.4) years old. Of patients providing data at all study time points, 37% (26/71) consistently used the portal prior to their hospital visits. The main reason for not completing PROMs was forgetting to do so in between visits (25/84, 29%). During an average session, patients viewed 5.3 different pages and spent 9.4 minutes logged on to the portal. Feedback from interviews was largely positive with no patients reporting difficulties navigating the survey and 50% of patients valuing the self-management tools provided in the portal. Regarding the portal content, patients were interested in reviewing their own results and reported high satisfaction with the dynamic self-management advice, also reflected in the high number of clicks on those pages. ConclusionsPatient portals can contribute to patient empowerment by offering sought-after information and self-management advice. In our study, the majority of our patients were open to using the portal. The low number of technical complaints and average time spent in the portal demonstrate the feasibility of our patient portal. While initial interest was high, long-term use was considerably lower and identified as the main area for improvement. In a next step, we will improve several aspects of the patient portal (eg, including a reminder to visit the portal before the next appointment and closer PROM symptom monitoring via an onconurse).

Multiple myeloma (MM) is a neoplasm affecting mainly elderly people. We developed a state-transition Markov model to project life expectancy (LE), lifetime costs, and incremental cost-effectiveness ratios (ICERs) of five sequential treatment options for elderly patients with MM in Austria. The treatment strategy "Starting with MPV" is likely to be a cost-effective treatment option for elderly patients with MM in Austria. [Extracted from the article]