Search

BackgroundDuring a pandemic, it is important for clinicians to stratify patients and decide who receives limited medical resources. Machine learning models have been proposed to accurately predict COVID-19 disease severity. Previous studies have typically tested only one machine learning algorithm and limited performance evaluation to area under the curve analysis. To obtain the best results possible, it may be important to test different machine learning algorithms to find the best prediction model. ObjectiveIn this study, we aimed to use automated machine learning (autoML) to train various machine learning algorithms. We selected the model that best predicted patients’ chances of surviving a SARS-CoV-2 infection. In addition, we identified which variables (ie, vital signs, biomarkers, comorbidities, etc) were the most influential in generating an accurate model. MethodsData were retrospectively collected from all patients who tested positive for COVID-19 at our institution between March 1 and July 3, 2020. We collected 48 variables from each patient within 36 hours before or after the index time (ie, real-time polymerase chain reaction positivity). Patients were followed for 30 days or until death. Patients’ data were used to build 20 machine learning models with various algorithms via autoML. The performance of machine learning models was measured by analyzing the area under the precision-recall curve (AUPCR). Subsequently, we established model interpretability via Shapley additive explanation and partial dependence plots to identify and rank variables that drove model predictions. Afterward, we conducted dimensionality reduction to extract the 10 most influential variables. AutoML models were retrained by only using these 10 variables, and the output models were evaluated against the model that used 48 variables. ResultsData from 4313 patients were used to develop the models. The best model that was generated by using autoML and 48 variables was the stacked ensemble model (AUPRC=0.807). The two best independent models were the gradient boost machine and extreme gradient boost models, which had an AUPRC of 0.803 and 0.793, respectively. The deep learning model (AUPRC=0.73) was substantially inferior to the other models. The 10 most influential variables for generating high-performing models were systolic and diastolic blood pressure, age, pulse oximetry level, blood urea nitrogen level, lactate dehydrogenase level, D-dimer level, troponin level, respiratory rate, and Charlson comorbidity score. After the autoML models were retrained with these 10 variables, the stacked ensemble model still had the best performance (AUPRC=0.791). ConclusionsWe used autoML to develop high-performing models that predicted the survival of patients with COVID-19. In addition, we identified important variables that correlated with mortality. This is proof of concept that autoML is an efficient, effective, and informative method for generating machine learning–based clinical decision support tools.

Fluorescence molecular tomography (FMT) is widely used in preclinical oncology research. FMT is the only imaging technique able to provide 3D distribution of fluorescent probes within thick highly scattering media. However, its integration into clinical medicine has been hampered by its low spatial resolution caused by the undetermined and ill-posed nature of its reconstruction algorithm. Another major factor degrading the quality of FMT images is the large backscattered excitation light component leaking through the rejection filters and coinciding with the weak fluorescent signal arising from a low tissue fluorescence concentration. In this paper, we present a new method based on the use of a novel thermo-sensitive fluorescence probe. In fact, the excitation light leakage is accurately estimated from a set of measurements performed at different temperatures and then is corrected for in the tomographic data. The obtained results show a considerable improvement in both spatial resolution and quantitative accuracy of FMT images due to the proper correction of fluorescent signals.

Alzheimer's disease (AD) is a devastating neurodegenerative disorder that impairs memory and causes cognitive and psychiatric deficits. New evidences indicate that AD is conceptualized as a disease of synaptic failure, although the molecular and cellular mechanisms underlying these defects remain to be elucidated. Determining the timing and nature of the early synaptic deficits is critical for understanding the progression of the disease and for identifying effective targets for therapeutic intervention. Using single-synapse functional and morphological analyses, we find that AMPA signaling, which mediates fast glutamatergic synaptic transmission in the central nervous system (CNS), is compromised early in the disease course in an AD mouse model. The decline in AMPA signaling is associated with changes in actin cytoskeleton integrity, which alters the number and the structure of dendritic spines. AMPA dysfunction and spine alteration correlate with the presence of soluble but not insoluble Aβ and tau species. In particular, we demonstrate that these synaptic impairments can be mitigated by Aβ immunotherapy. Together, our data suggest that alterations in AMPA signaling and cytoskeletal processes occur early in AD. Most important, these deficits are prevented by Aβ immunotherapy, suggesting that existing therapies, if administered earlier, could confer functional benefits.

Purpose Bronchial brushings (BB) commonly aid in the diagnosis of primary lung cancer. However, the utility of this method in diagnosing endobronchial metastases (EBM) from extrapulmonic malignancies has not been thoroughly evaluated. The purpose of this study is to evaluate the sensitivity of BB in diagnosing EBM. Methods An institutional database was queried for all patients with cytologically or histologically confirmed extrapulmonary EBM identified by endobronchial biopsy between 1978 and 2013. Data were collected on patient demographics, histologic and cytologic diagnoses, time from primary malignancy to identification of EBM, and location of EBM. The sensitivity of BB for the diagnosis of EBM and the clinicopathologic features of extrapulmonary EBM were assessed. Results Fifty-six patients (33 females, 23 males; mean age 53 years) were identified with EBM. Diagnoses included lymphoma (21), breast adenocarcinoma (11), colonic adenocarcinoma (7), melanoma (6), renal cell carcinoma (RCC, 5), embryonal carcinoma (2), and 1 case each of tonsillar squamous cell carcinoma, thymic carcinoma, leiomyosarcoma, and sarcoma, not otherwise specified. The sensitivity of BB for identifying EBM was 85% overall and 94% for non-hematologic malignancies. The mean interval between primary diagnosis and EBM was 59 months (range 0-264 months). Excluding ten patients who had EBM at their initial presentation, lymphoma had the shortest (10 months) and RCC had the longest (264 months) mean interval between primary diagnosis and EBM. The mean time between EBM identification and death was 22.4 months (n = 24). Conclusion Bronchial brushing is a sensitive technique for diagnosing non-hematologic extrapulmonic endobronchial metastases., Author(s): Kenji Ikemura [sup.1] [sup.3] , Diana Murro Lin [sup.1] , Colin P. Martyn [sup.1] , Ji-Weon Park [sup.1] , Christopher W. Seder [sup.2] , Paolo Gattuso [sup.1] Author Affiliations: [...]

A 44-year-old man with a history of renal transplantation presented with right lower abdominal wall swelling, redness and pain. A bacterial abscess was drained, and he was discharged home with oral antibiotics. After failing to improve, he returned to the hospital, where he was briefly treated with intravenous antibiotics and discharged home again. The patient returned 5 days later, reporting worsening right groin swelling that extended into the ipsilateral scrotum. Imaging revealed a persistent fluid collection in the region, and he was taken for surgical debridement. Tissue immunochemistry and histopathological evaluation identified cytomegalovirus infection. Plasma quantitative PCR for cytomegalovirus demonstrated high viraemia. The patient was successfully treated with intravenous ganciclovir, followed by oral valganciclovir, with resolution of the skin changes. Persistent hydrocele with epididymitis on imaging suggests that this process may have been the source of the cutaneous cytomegalovirus infection. [ABSTRACT FROM AUTHOR]

Tacrolimus is important for immunosuppression in kidney transplantation. In this historical cohort and in vitro study, we evaluated the changes in tacrolimus pharmacokinetics early after living donor kidney transplantation and the effects of interleukin (IL)‐6 on cytochrome P450 3A4 (CYP3A4) and cytochrome P450 3A5 (CYP3A5) expression. In the historical cohort study, 22 patients who met the inclusion criteria were classified into CYP3A5 expressors and non‐expressors (n = 16 and 6, respectively). The blood tacrolimus concentration per dose ratio (C/D) temporarily increased post–kidney transplantation on days 3‐4 only in CYP3A5 non‐expressors. The effects of IL‐6 on CYP3A4 and CYP3A5 expression were also investigated in vitro using HepG2 and Caco‐2 cells. IL‐6 induced a significant concentration‐ and time‐dependent decrease in CYP3A4 and CYP3A5 expression in both cells. The mean CYP3A4 expression level at 12 hours after IL‐6 exposure (% of 0 hour) was 44.0 and 62.6 in HepG2 and Caco‐2 cells, respectively, whereas the CYP3A5 expression level was 30.7 and 52.4, respectively. We hypothesize that CYP3A5 non‐expressors might exhibit a temporary decrease in the oral clearance of tacrolimus via an increase in serum IL‐6 concentrations early after kidney transplantation. These results may help develop strategies to improve kidney transplant outcome. [ABSTRACT FROM AUTHOR]