Whole-Slide Cytometric Feature Mapping for Distinguishing Tumor Genomic Subtypes in Head and Neck Squamous Cell Carcinoma Whole-Slide Images.

Head and neck squamous cell carcinoma (HNSCC) is a heterogeneous disease where, in advanced stages, clinical and pathologic stages do not correlate with outcome. Molecular and genomic biomarkers for HNSCC classification have shown promise for prognostic and therapeutic applications. This study utilized automated image analysis techniques in whole-slide images of HNSCC tumors to identify relationships between cytometric features and genomic phenotypes. Hematoxylin and eosin-stained slides of HNSCC tumors (N = 49) were obtained from The Cancer Imaging Archive, along with accompanying clinical, pathologic, genomic, and proteomic reports. Automated nuclear detection was performed across the entirety of slides, and cytometric feature maps were generated. Forty-one cytometric features were evaluated for associations with tumor grade, tumor stage, tumor subsite, and integrated genomic subtype. Thirty-two features demonstrated significant association with integrated genomic subtype when corrected for multiple comparisons. In particular, the basal subtype was visually distinguishable from the chromosomal instability and immune subtypes based on cytometric feature measurements. No features were significantly associated with tumor grade, stage, or subsite. This study provides preliminary evidence that features derived from tissue pathology slides could provide insights into genomic phenotypes of HNSCC.

NeoConnect: The Design, Implementation, and Impact of a Virtual Family-Centered NICU Program.

BACKGROUND: Familial involvement in the neonatal intensive care unit (NICU) reduces parental stress and strengthens parental-infant bonding. However, parents often face barriers to in-person visitation. The coronavirus disease-2019 COVID-19 pandemic has exacerbated limitations to parental bedside presence. OBJECTIVE: To design, implement, and evaluate a technology-based program to connect NICU babies with their families during the COVID-19 pandemic. METHODS: We created NeoConnect at our level IV NICU, which included parental audio recordings and video chats between parents and their babies. Parental and NICU staff input on NeoConnect was gathered via preimplementation surveys. Inaugural families and staff members completed a postparticipation survey. RESULTS: Prior to implementation, all parents who were surveyed (n = 24) wished they could be more involved in their baby's care. In the first 3 months of NeoConnect, 48 families participated in the audio recording project and 14 families participated in the video chat initiative. Following implementation, 85% of surveyed staff (28/33) reported that the patients became calmer when listening to their parents' recorded voice and 100% of surveyed parents (6/6) reported that video chats reduced their stress level. CONCLUSION: Harnessing technology as a tool to increase parental involvement in the NICU is feasible and beneficial for NICU patients and their families.

Deep learning for survival outcomes.

Deep learning is a class of machine learning algorithms that are popular for building risk prediction models. When observations are censored, the outcomes are only partially observed and standard deep learning algorithms cannot be directly applied. We develop a new class of deep learning algorithms for outcomes that are potentially censored. To account for censoring, the unobservable loss function used in the absence of censoring is replaced by a censoring unbiased transformation. The resulting class of algorithms can be used to estimate both survival probabilities and restricted mean survival. We show how the deep learning algorithms can be implemented by adapting software for uncensored data by using a form of response transformation. We provide comparisons of the proposed deep learning algorithms to existing risk prediction algorithms for predicting survival probabilities and restricted mean survival through both simulated datasets and analysis of data from breast cancer patients.

Glacial Acetic Acid Adverse Events: Case Reports and Review of the Literature.

Glacial acetic acid is a dangerous chemical that has been associated with several adverse drug events involving patients over recent years. When diluted to the proper concentration, acetic acid solutions have a variety of medicinal uses. Unfortunately, despite warnings, the improper dilution of concentrated glacial acetic acid has resulted in severe burns and other related morbidities. We report on 2 additional case reports of adverse drug events involving glacial acetic acid as well as a review of the literature. A summary of published case reports is provided, including the intended and actual concentration of glacial acetic acid involved, the indication for use, degree of exposure, and resultant outcome. Strategies that have been recommended to improve patient safety are summarized within the context of the key elements of the medication use process.

Assessment of the Diagnostic Performance of a Commercially Available Artificial Intelligence Algorithm for Risk Stratification of Thyroid Nodules on Ultrasound.

Background: Thyroid nodules are challenging to accurately characterize on ultrasound (US), though the emergence of risk stratification systems and more recently artificial intelligence (AI) algorithms has improved nodule classification. The purpose of this study was to evaluate the performance of a recent Food and Drug Administration (FDA)-cleared AI tool for detection of malignancy in thyroid nodules on US. Methods: One year of consecutive thyroid US with ≥1 nodule from Duke University Hospital and its affiliate community hospital (649 nodules from 347 patients) were retrospectively evaluated. Included nodules had ground truth diagnoses by surgical pathology, fine needle aspiration (FNA), or three-year follow-up US showing stability. An FDA-cleared AI tool (Koios DS Thyroid) analyzed each nodule to generate (i) American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) descriptors, scores, and follow-up recommendations and (ii) an AI-adapter score to further adjust risk assessments and recommendations. Four groups were then compared: (i) Koios with AI-adapter, (ii) Koios without AI-adapter, (iii) clinical radiology report, and (iv) radiology report combined with AI-adapter. Performance of the final recommendations (FNA or no FNA) was determined based on ground truth, and comparison between the four groups was made using sensitivity, specificity, and receiver-operating-curve analysis. Results: Of 649 nodules, 32 were malignant and 617 were benign. Performance of Koios with AI-adapter enabled was similar to radiologists (area under the curve [AUC] 0.70 for both, [CI 0.60-0.81] and [0.60-0.79], respectively). Koios with AI-adapter had improved specificity compared to radiologists (0.63 [CI: 0.59-0.67] versus 0.43 [CI: 0.38-0.48]) but decreased sensitivity (0.69 [CI: 0.50-0.83) versus 0.81 [CI: 0.61, 0.92]). Highest performance was seen when the radiology interpretation was combined with the AI-adapter (AUC 0.76, [CI: 0.67-0.85]). Combined with the AI-adapter, radiologist specificity improved from 0.43 ([CI: 0.38-0.48]) to 0.53 ([CI: 0.49-0.58]) (McNemar's test p < 0.001), resulting in 17% fewer FNA recommendations, with unchanged sensitivity (0.81, p = 1). Conclusion: Koios DS demonstrated standalone performance similar to radiologists, though with lower sensitivity and higher specificity. Performance was best when radiologist interpretations were combined with the AI-adapter component, with improved specificity and reduced unnecessary FNA recommendations.

Effects of color variation and physiological state on ascidian microbiomes.

Ascidians, known for their color variation, host species-specific microbial symbiont communities. Some ascidians can also transition into a nonfiltering (resting) physiological state. Recent studies suggest that the microbial symbiont communities may vary across different physiological states and color morphs of the host. The colonial ascidian, Polyclinum constellatum, which exhibits several color morphs in the Caribbean Sea, periodically ceases its filtering activity. To investigate if color variation in P. constellatum is indicative of sibling speciation, we sequenced fragments of the ribosomal 18S rRNA and the mitochondrial cytochrome oxidase subunit I genes. Additionally, we sequenced a fragment of the 16S rRNA gene to characterize the microbial communities of two common color morphs (red and green) in colonies that were either actively filtering (active) or nonfiltering (resting). Phylogenetic analyses of both ascidian genes resulted in well-supported monophyletic clades encompassing all color variants of P. constellatum. Interestingly, no significant differences were observed among the microbial communities of the green and red morphs, suggesting that color variation in this species is a result of intraspecific variation. However, the host's physiological state significantly influenced the microbial community structure. Nonfiltering (resting) colonies hosted higher relative abundances of Kiloniella (Alphaproteobacteria) and Fangia (Gammaproteobacteria), while filtering colonies hosted more Reugeria (Alphaproteobacteria) and Endozoicomonas (Gammaproteobacteria). This study demonstrates that microbial symbiont communities serve as reliable indicators of the taxonomic state of their host and are strongly influenced by the host's feeding condition.

Diagnostic Performance of Renal Contrast Excretion on Early Phase CT Myelography in Spontaneous Intracranial Hypotension.

BACKGROUND AND PURPOSE: Early opacification of the renal collecting system during CT myelography (CTM) performed for the evaluation of Spontaneous Intracranial Hypotension (SIH) has been demonstrated in prior studies. However, these investigations often included CTMs scanned >30 minutes after intrathecal contrast injection, a longer delay than the myelographic techniques used in current practice. The purpose of this study was to determine whether renal contrast excretion (RCE) measured during this earlier time period (≤30 minutes) can discriminate patients with SIH from patients without SIH. MATERIALS AND METHODS: Single-center, retrospective cohort of consecutive patients presenting for evaluation of possible SIH between July 2021-May 2022. RCE was measured in both renal hila using standardized (5-15mm3) ROIs. ROC curves were constructed comparing RCE between patients with SIH to patients without SIH in the overall cohort, and within the subgroup of patients with negative myelograms. RESULTS: The study cohort included 190 subjects. Both unadjusted and adjusted models demonstrated a statistically significant increase in renal contrast density among patients with SIH compared to those without SIH (p-values ≤ 0.001). The ROC curve showed moderate discrimination between these groups (AUC 0.76). However, using clinically meaningful test criteria of sensitivity >90% or specificity >90%, the two corresponding threshold HU values resulted in low specificity of 31.3% and sensitivity of 50.8%. Subgroup analysis of patients with negative myelograms showed poorer performance in discriminating SIH+ from SIH- (AUC 0.62). In this subgroup, using similar test criteria of sensitivity >90% or specificity >90 resulted in low specificities and sensitivities, at 26.0% and 37.5% respectively. CONCLUSIONS: We found a statistically significant positive association between RCE and SIH diagnosis during early-phase CTM, however clinically useful thresholds based on cutoff values for renal HU resulted in poor sensitivities or specificities, with substantial false positives or false negatives, respectively. Thus, while we confirmed statistically significant differences in RCE in the ≤30 min time period, in keeping with prior investigations of more delayed time periods, overlap in renal attenuation values prevented the development of clinically useful threshold value for discriminating SIH+ from SIH-patients. ABBREVIATIONS: SIH = spontaneous intracranial hypotension; RCE = renal contrast excretion; CTM = CT myelography; CVF = CSF-venous fistula; ICHD-3 = international classification of headache disorders third edition; CKD = chronic kidney disease.

Subventricular zone stem cell niche injury is associated with intestinal perforation in preterm infants and predicts future motor impairment.

Brain injury is highly associated with preterm birth. Complications of prematurity, including spontaneous or necrotizing enterocolitis (NEC)-associated intestinal perforations, are linked to lifelong neurologic impairment, yet the mechanisms are poorly understood. Early diagnosis of preterm brain injuries remains a significant challenge. Here, we identified subventricular zone echogenicity (SVE) on cranial ultrasound in preterm infants following intestinal perforations. The development of SVE was significantly associated with motor impairment at 2 years. SVE was replicated in a neonatal mouse model of intestinal perforation. Examination of the murine echogenic subventricular zone (SVZ) revealed NLRP3-inflammasome assembly in multiciliated FoxJ1+ ependymal cells and a loss of the ependymal border in this postnatal stem cell niche. These data suggest a mechanism of preterm brain injury localized to the SVZ that has not been adequately considered. Ultrasound detection of SVE may serve as an early biomarker for neurodevelopmental impairment after inflammatory disease in preterm infants.

Survival analysis using deep learning with medical imaging.

There is widespread interest in using deep learning to build prediction models for medical imaging data. These deep learning methods capture the local structure of the image and require no manual feature extraction. Despite the importance of modeling survival in the context of medical data analysis, research on deep learning methods for modeling the relationship of imaging and time-to-event data is still under-developed. We provide an overview of deep learning methods for time-to-event outcomes and compare several deep learning methods to Cox model based methods through the analysis of a histology dataset of gliomas.

Ability of the respiratory ECMO survival prediction (RESP) score to predict survival for patients with COVID-19 ARDS and non-COVID-19 ARDS: a single-center retrospective study.

The respiratory ECMO survival prediction (RESP) score is used to predict survival for patients managed with extracorporeal membrane oxygenation (ECMO), but its performance in patients with Coronavirus Disease 2019 (COVID-19) acute respiratory distress syndrome (ARDS) is unclear. We evaluated the ability of the RESP score to predict survival for patients with both non-COVID 19 ARDS and COVID-19 ARDS managed with ECMO at our institution. Receiver operating characteristic area under the curve (AUC) analysis found the RESP score reasonably predicted survival in patients with non-COVID-19 ARDS (AUC 0.76, 95% CI 0.68-0.83), but not patients with COVID-19 ARDS (AUC 0.54, 95% CI 0.41-0.66).

U-Shaped Association Between Carboxyhemoglobin and Mortality in Patients With Acute Respiratory Distress Syndrome on Venovenous Extracorporeal Membrane Oxygenation.

Background: Carbon monoxide (CO) is an endogenous signaling molecule that activates cytoprotective programs implicated in the resolution of acute respiratory distress syndrome (ARDS) and survival of critical illness. Because CO levels can be measured in blood as carboxyhemoglobin, we hypothesized that carboxyhemoglobin percent (COHb%) may associate with mortality. OBJECTIVES: To examine the relationship between COHb% and outcomes in patients with ARDS requiring venovenous extracorporeal membrane oxygenation (ECMO), a condition where elevated COHb% is commonly observed. DESIGN: Retrospective cohort study. SETTING: Academic medical center ICU. PATIENTS: Patients were included that had ARDS on venovenous ECMO. MEASUREMENTS AND MAIN RESULTS: We examined the association between COHb% and mortality using a Cox proportional hazards model. Secondary outcomes including ECMO duration, ventilator weaning, and hospital and ICU length of stay were examined using both subdistribution and causal-specific hazard models for competing risks. We identified 109 consecutive patients for analysis. Mortality significantly decreased per 1 U increase in COHb% below 3.25% (hazard ratio [HR], 0.35; 95% CI, 0.15-0.80; p = 0.013) and increased per 1 U increase above 3.25% (HR, 4.7; 95% CI, 1.5-14.7; p = 0.007) reflecting a nonlinear association (p = 0.006). Each unit increase in COHb% was associated with reduced likelihood of liberation from ECMO and mechanical ventilation, and increased time to hospital and ICU discharge (all p < 0.05). COHb% was significantly associated with hemolysis but not with initiation of hemodialysis or blood transfusions. CONCLUSIONS: In patients with ARDS on venovenous ECMO, COHb% is a novel biomarker for mortality exhibiting a U-shaped pattern. Our findings suggest that too little CO (perhaps due to impaired host signaling) or excess CO (perhaps due to hemolysis) is associated with higher mortality. Patients with low COHb% may exhibit the most benefit from future therapies targeting anti-oxidant and anti-inflammatory pathways such as low-dose inhaled CO gas.

Racial Disparities in Lung Cancer Screening Among Veterans, 2013 to 2021.

Importance: Racial disparities in lung cancer screening (LCS) are often ascribed to barriers such as cost, insurance status, access to care, and transportation. Because these barriers are minimized within the Veterans Affairs system, there is a question of whether similar racial disparities exist within a Veterans Affairs health care system in North Carolina. Objectives: To examine whether racial disparities in completing LCS after referral exist at the Durham Veterans Affairs Health Care System (DVAHCS) and, if so, what factors are associated with screening completion. Design, Setting, and Participants: This cross-sectional study assessed veterans referred to LCS between July 1, 2013, and August 31, 2021, at the DVAHCS. All included veterans self-identified as White or Black and met the US Preventive Services Task Force eligibility criteria as of January 1, 2021. Participants who died within 15 months of consultation or who were screened before consultation were excluded. Exposures: Self-reported race. Main Outcomes and Measures: Screening completion was defined as completing computed tomography for LCS. The associations among screening completion, race, and demographic and socioeconomic risk factors were assessed using logistic regression models. Results: A total of 4562 veterans (mean [SD] age, 65.4 [5.7] years; 4296 [94.2%] male; 1766 [38.7%] Black and 2796 [61.3%] White) were referred for LCS. Of all veterans referred, 1692 (37.1%) ultimately completed screening; 2707 (59.3%) never connected with the LCS program after referral and an informational mailer or telephone call, indicating a critical point in the LCS process. Screening rates were substantially lower among Black compared with White veterans (538 [30.5%] vs 1154 [41.3%]), with Black veterans having 0.66 times lower odds (95% CI, 0.54-0.80) of screening completion after adjusting for demographic and socioeconomic factors. Conclusions and Relevance: This cross-sectional study found that after referral for initial LCS via a centralized program, Black veterans had 34% lower odds of LCS screening completion compared with White veterans, a disparity that persisted even after accounting for numerous demographic and socioeconomic factors. A critical point in the screening process was when veterans must connect with the screening program after referral. These findings may be used to design, implement, and evaluate interventions to improve LCS rates among Black veterans.

Age-related changes to adipose tissue and peripheral neuropathy in genetically diverse HET3 mice differ by sex and are not mitigated by rapamycin longevity treatment.

Neural communication between the brain and adipose tissues regulates energy expenditure and metabolism through modulation of adipose tissue functions. We have recently demonstrated that under pathophysiological conditions (obesity, diabetes, and aging), total subcutaneous white adipose tissue (scWAT) innervation is decreased ('adipose neuropathy'). With advanced age in the C57BL/6J mouse, small fiber peripheral nerve endings in adipose tissue die back, resulting in reduced contact with adipose-resident blood vessels and other cells. This vascular neuropathy and parenchymal neuropathy together likely pose a physiological challenge for tissue function. In the current work, we used the genetically diverse HET3 mouse model to investigate the incidence of peripheral neuropathy and adipose tissue dysregulation across several ages in both male and female mice. We also investigated the anti-aging treatment rapamycin, an mTOR inhibitor, as a means to prevent or reduce adipose neuropathy. We found that HET3 mice displayed a reduced neuropathy phenotype compared to inbred C56BL/6 J mice, indicating genetic contributions to this aging phenotype. Compared to female HET3 mice, male HET3 mice had worse neuropathic phenotypes by 62 weeks of age. Female HET3 mice appeared to have increased protection from neuropathy until advanced age (126 weeks), after reproductive senescence. We found that rapamycin overall had little impact on neuropathy measures, and actually worsened adipose tissue inflammation and fibrosis. Despite its success as a longevity treatment in mice, higher doses and longer delivery paradigms for rapamycin may lead to a disconnect between life span and beneficial health outcomes.

Microplastics in Tampa Bay, Florida: Abundance and variability in estuarine waters and sediments.

This study provides the first measurement of microplastic abundance and distribution in surface waters and sediments in Tampa Bay, FL. Microplastic concentrations in discrete water samples ranged from 0.25 to 7.0 particles/L with an average of 0.94 (±0.52) particles/L. Samples taken with a 330 µm plankton net had 1.2-18.1 particles/m3 with an average of 4.5 (±2.3) particles/m3. Discrete samples were 200 times higher than net samples, suggesting substantial losses or undersampling with the net. For both discrete and plankton tow samples, there were no significant differences in concentrations between stations or regions. Intense rainfall events in the summer always preceded samples with substantially higher counts. Most (>75%) microplastics were fibers. Using an average value of 1 particle/L, Tampa Bay contains ~4 billion microplastic particles. Surface sediments had an average of 280 (±290) particles/kg, ranging from 30 to 790 particles/kg. Highest concentrations of microplastics were found in sediments close to industrial sources; lowest values in Middle and Lower Tampa Bay are consistent with shorter residence times.

The role of the medial prefrontal cortex in social categorization.

Group membership is an important aspect of our everyday behavior. Recently, we showed that existing relevant in-group labels increased activation in the medial prefrontal cortex (MPFC) compared with out-group labels, suggesting a role of the MPFC in social categorization. However, the question still remains whether this increase in MPFC activation for in-group representation is solely related with previous experience with the in-group. To test this, we randomly assigned participants to a red or blue team and in a subsequent functional magnetic resonance imaging experiment they categorized red and blue team words as belonging to either the in-group or the out-group. Results showed that even under these minimal conditions increased activation was found in the MPFC when participants indicated that they belonged to a group, as compared with when they did not. This effect was found to be associated with the level of group identification. These results confirm the role of MPFC in social categorization.

The neuroscience of group membership.

The present study aimed to uncover the neural activity associated with specific in-group and out-group word related stimuli, to examine the neuroanatomical basis of group membership concept representation, and investigate to what extent neural processes represent 'in-group' differently from 'out-group'. Participants' brain activity was measured with functional MRI while they had to categorize social, in-group and out-group words and non-social, living and non-living words. The results showed that a network of brain regions previously identified as the 'social brain', including the cortical midline structures, tempo-parietal junction and the anterior temporal gyrus showed enhanced activation for social words versus non-social words. Crucially, the processing of in-group words compared to the out-group words activated a specific network including the ventral medial prefrontal and anterior and dorsal cingulate cortex. These regions correspond to a neural network previously identified as the 'personal self'. Our results suggest that the 'social' and 'personal self' are closely related and that we derive our self image from the groups we belong to.

Group cognitive-behavioral therapy for hoarding disorder: an open trial.

Although cognitive-behavioral therapy (CBT) appears to be a promising treatment approach for hoarding disorder, treatment to date has been quite labor intensive. The goal of this study, therefore, was to assess the potential effectiveness of group CBT for hoarding, without home visits by the clinician. Forty-five individuals with hoarding disorder enrolled in either a 16 or 20 session program of group CBT; 30 (67%) completed treatment. Using mixed-effects models to account for missing data, we report data from 35 (78%) participants who provided enough data for analysis. Participants demonstrated significant improvements in hoarding symptoms, as well as symptoms of depression and anxiety, and quality of life. Improvements in hoarding symptoms were comparable to two published clinical trials on individual CBT for hoarding disorder. Results of this study suggest that group CBT for hoarding, without home discarding sessions by the clinician, may be an effective treatment option with the potential advantage of increasing treatment access by reducing clinician burden and cost of treatment.