Citations

Background and motivation: Lung computed tomography (CT) techniques are high-resolution and are well adopted in the intensive care unit (ICU) for COVID-19 disease control classification. Most artificial intelligence (AI) systems do not undergo generalization and are typically overfitted. Such trained AI systems are not practical for clinical settings and therefore do not give accurate results when executed on unseen data sets. We hypothesize that ensemble deep learning (EDL) is superior to deep transfer learning (TL) in both non-augmented and augmented frameworks. Methodology: The system consists of a cascade of quality control, ResNet–UNet-based hybrid deep learning for lung segmentation, and seven models using TL-based classification followed by five types of EDL’s. To prove our hypothesis, five different kinds of data combinations (DC) were designed using a combination of two multicenter cohorts—Croatia (80 COVID) and Italy (72 COVID and 30 controls)—leading to 12,000 CT slices. As part of generalization, the system was tested on unseen data and statistically tested for reliability/stability. Results: Using the K5 (80:20) cross-validation protocol on the balanced and augmented dataset, the five DC datasets improved TL mean accuracy by 3.32%, 6.56%, 12.96%, 47.1%, and 2.78%, respectively. The five EDL systems showed improvements in accuracy of 2.12%, 5.78%, 6.72%, 32.05%, and 2.40%, thus validating our hypothesis. All statistical tests proved positive for reliability and stability. Conclusion: EDL showed superior performance to TL systems for both (a) unbalanced and unaugmented and (b) balanced and augmented datasets for both (i) seen and (ii) unseen paradigms, validating both our hypotheses.

5554 Background: Epithelial ovarian cancer (EOC) is characterized by chemoresistant recurrences primarily confined to the intraperitoneal cavity, frequently leading to malignant small bowel obstructions. Intraperitoneal (IP) chemotherapy has been incorporated in EOC, but only in the first-line setting. Pressurized intraperitoneal aerosolized chemotherapy (PIPAC) is a novel IP drug delivery method that optimizes tissue penetration depth and drug distribution, to treat recurrent peritoneal malignancies. Recent trials outside the U.S. demonstrated partial to stable response following PIPAC in recurrent EOC. To date, no clinical trials have been completed investigating the role of PIPAC in EOC in the United States. Methods: We performed a multicenter prospective phase I trial using IP cisplatin 10.5 mg/m2 and doxorubicin 2.1 mg/m2 via PIPAC at 6-week intervals. Patients were eligible if they had recurrent disease with peritoneal metastasis; extraperitoneal metastasis was allowed. The primary endpoint was dose-limiting toxicity (DLT). Secondary endpoints included progression free survival, PCI scoring, RECIST measurements, and quality of life (QoL). The patients with gastric and uterine cancers’ toxicity data were used but given inherent prognostic differences the treatment efficacy for these patients were not used in this analysis. Adverse event (AE) monitoring was graded using CTCAE spanning 18 weeks. Following completion of treatment (³2 PIPACs) patients were followed up every 12 weeks. QoL was measured using patient-reported assessments at 0, 6, 12, and 18 weeks. Baseline demographics were summarized. Mean change in PCI was calculated between cycle (C) 1 and 2, and cycle 2 and 3. Toxicities attributed to the surgery or PIPAC were reported with highest grade for each patient. Results: Nine heavily pretreated, recurrent patients were enrolled in the trial: 7 with platinum-resistant recurrent EOC, 1 with uterine cancer, and 1 with gastric cancer. PIPAC completion rate was 71.4%. Median age of participants was 65 years. Median prior lines of chemotherapy were 4 (range 2 - 10). No surgical complications occurred in the cohort. Two patients came off study: one for G3 abdominal pain and the other for G3 anorexia. For the EOC cohort, the median PCI reduction was -1.4% (C1 to C2) and -2.8% (C1 to C3). One low-grade serous EOC patient demonstrated a partial response based on RECIST criteria and received 6 cycles. Two EOC patients showed an initial response based on PCI reduction, but progressed following Cycle 2. Overall, QoL was stable. Conclusions: PIPAC with cisplatin/doxorubicin in platinum-resistant ovarian cancer is well tolerated. Intraperitoneal responses were seen in a subset of low-grade serous ovarian cancer patients, which may warrant further study. Heavily pretreated, platinum-resistant high-grade EOC patients derived limited intraperitoneal control from PIPAC monotherapy. Clinical trial information: NCT04329494 .

BACKGROUND AND OBJECTIVES: Surgical approaches to the ventral brainstem and petroclival regions are complex, and standard retrosigmoid and subtemporal approaches are often inadequate. Retrosigmoid suprameatal tubercle (RSMTA) and anterior subtemporal transpetrosal (ASTA) approaches may provide extended surgical exposure with less brain retraction. The objective of this study was to evaluate advantages and disadvantages of RSMTA vs ASTA, and illustrate the surgical corridors and 3-dimensional microsurgical anatomy of the related structures. METHODS: Four silicone-injected adult cadaver heads (8 sides) were dissected to evaluate the accessibility of lesions located at the petrous apex, ventral brainstem, and pontomedullary region using ASTA and RSTMA. RESULTS: Both ASTA and RSMTA provide access from the petrous apex to the ventral lower pons and pontomedullary junction. A greater extent of safely resected bone was found in ASTA vs RSMTA. ASTA provides a larger surgical view to the ventrolateral midpons, peritrigeminal region, superior neurovascular complex, pontomesencephalic junction, and posterior cavernous sinus. Meanwhile, through cranial nerve V mobilization, RSMTA provides a larger surgical view to the lower half part of the pons, ventrolateral part of the pontomedullary junction, and middle and lower neurovascular structures. CONCLUSION: The choice of surgical approach is determined by considering the area where the lesion originates, lesion size, the anatomic structures to which it extends, and evaluation of the area that can be surgically exposed. Our study highlights the differences between these approaches and important surgical anatomic considerations.

Acute myeloid leukemia (AML) is a heterogeneous hematologic malignancy characterized by the clonal expansion of myeloid blasts in the peripheral blood, bone marrow, and/or other tissues. It is the most common form of acute leukemia among adults and accounts for the largest number of annual deaths from leukemias in the United States. Like AML, blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a myeloid malignancy. It is a rare malignancy characterized by the aggressive proliferation of precursors of plasmacytoid dendritic cells that frequently involves the bone marrow, skin, central nervous system, and other organs and tissues. This discussion section focuses on the diagnosis and management of BPDCN as outlined in the NCCN Guidelines for AML.

AbstractBackgroundAortic stenosis (AS) is associated with myocardial ischemia through different mechanisms and may impair coronary arterial flow. However, data on the impact of moderate AS in patients with acute myocardial infarction (MI) is limited.AimsThis study aimed to investigate the impact of moderate AS in patients presenting with acute myocardial infarction (MI).MethodsWe conducted a retrospective analysis of all patients who presented with acute MI to all Mayo Clinic hospitals, using the Enterprise Mayo PCI Database from 2005 to 2016. Patients were stratified into two groups: moderate AS and mild/no AS. The primary outcome was all cause mortality.ResultsThe moderate AS group included 183 (13.3%) patients, and the mild/no AS group included 1190 (86.7%) patients. During hospitalization, there was no difference between both groups in mortality. Patients with moderate AS had higher in‐hospital congestive heart failure (CHF) (8.2% vs. 4.4%, p = 0.025) compared with mild/no AS patients. At 1‐year follow‐up, patients with moderate AS had higher mortality (23.9% vs. 8.1%, p < 0.001) and higher CHF hospitalization (8.3% vs. 3.7%, p = 0.028). In multivariate analysis, moderate AS was associated with higher mortality at 1‐year (odds ratio 2.4, 95% confidence interval [1.4–4.1], p = 0.002). In subgroup analyses, moderate AS increased all‐cause mortality in STEMI and NSTEMI patients.ConclusionThe presence of moderate AS in acute MI patients was associated with worse clinical outcomes during hospitalization and at 1‐year follow‐up. These unfavorable outcomes highlight the need for a close follow‐up of these patients and for timely therapeutic strategies to best manage these coexisting conditions.

This study reports the needs-based development, effectiveness and feasibility of a novel, comprehensive spinal cord stimulation (SCS) digital curriculum designed for pain medicine trainees. The curriculum aims to address the documented systematic variability in SCS education and empower physicians with SCS expertise, which has been linked to utilization patterns and patient outcomes. Following a needs assessment, the authors developed a three-part SCS e-learning video curriculum with baseline and postcourse knowledge tests. Best practices were used for educational video production and test-question development. The study period was from 1 February 2020 to 31 December 2020. A total of 202 US-based pain fellows across two cohorts (early-fellowship and late-fellowship) completed the baseline knowledge assessment, while 122, 96 and 88 participants completed all available post-tests for Part I (Fundamentals), Part II (Cadaver Lab) and Part III (Decision Making, The Literature and Critical Applications), respectively. Both cohorts significantly increased knowledge scores from baseline to immediate post-test in all curriculum parts (p<0.001). The early-fellowship cohort experienced a higher rate of knowledge gain for Parts I and II (p=0.045 and p=0.027, respectively). On average, participants viewed 6.4 out of 9.6 hours (67%) of video content. Self-reported prior SCS experience had low to moderate positive correlations with Part I and Part III pretest scores (r=0.25, p=0.006; r=0.37, p<0.001, respectively). Initial evidence suggests that Pain Rounds provides an innovative and effective solution to the SCS curriculum deficit. A future controlled study should examine this digital curriculum’s long-term impact on SCS practice and treatment outcomes.

OBJECTIVES: To develop evidence-based recommendations for clinicians caring for adults with acute liver failure (ALF) or acute on chronic liver failure (ACLF) in the ICU. DESIGN: The guideline panel comprised 27 members with expertise in aspects of care of the critically ill patient with liver failure or methodology. We adhered to the Society of Critical Care Medicine standard operating procedures manual and conflict-of-interest policy. Teleconferences and electronic-based discussion among the panel, as well as within subgroups, served as an integral part of the guideline development. INTERVENTIONS: In part 2 of this guideline, the panel was divided into four subgroups: neurology, peri-transplant, infectious diseases, and gastrointestinal groups. We developed and selected Population, Intervention, Comparison, and Outcomes (PICO) questions according to importance to patients and practicing clinicians. For each PICO question, we conducted a systematic review and meta-analysis where applicable. The quality of evidence was assessed using the Grading of Recommendations Assessment, Development, and Evaluation approach. We used the evidence to decision framework to facilitate recommendations formulation as strong or conditional. We followed strict criteria to formulate best practice statements. MEASUREMENTS AND MAIN RESULTS: We report 28 recommendations (from 31 PICO questions) on the management ALF and ACLF in the ICU. Overall, five were strong recommendations, 21 were conditional recommendations, two were best-practice statements, and we were unable to issue a recommendation for five questions due to insufficient evidence. CONCLUSIONS: Multidisciplinary, international experts formulated evidence-based recommendations for the management ALF and ACLF patients in the ICU, acknowledging that most recommendations were based on low quality and indirect evidence.