TMTV-Net: fully automated total metabolic tumor volume segmentation in lymphoma PET/CT images - a multi-center generalizability analysis.

PURPOSE: Total metabolic tumor volume (TMTV) segmentation has significant value enabling quantitative imaging biomarkers for lymphoma management. In this work, we tackle the challenging task of automated tumor delineation in lymphoma from PET/CT scans using a cascaded approach. METHODS: Our study included 1418 2-[18F]FDG PET/CT scans from four different centers. The dataset was divided into 900 scans for development/validation/testing phases and 518 for multi-center external testing. The former consisted of 450 lymphoma, lung cancer, and melanoma scans, along with 450 negative scans, while the latter consisted of lymphoma patients from different centers with diffuse large B cell, primary mediastinal large B cell, and classic Hodgkin lymphoma cases. Our approach involves resampling PET/CT images into different voxel sizes in the first step, followed by training multi-resolution 3D U-Nets on each resampled dataset using a fivefold cross-validation scheme. The models trained on different data splits were ensemble. After applying soft voting to the predicted masks, in the second step, we input the probability-averaged predictions, along with the input imaging data, into another 3D U-Net. Models were trained with semi-supervised loss. We additionally considered the effectiveness of using test time augmentation (TTA) to improve the segmentation performance after training. In addition to quantitative analysis including Dice score (DSC) and TMTV comparisons, the qualitative evaluation was also conducted by nuclear medicine physicians. RESULTS: Our cascaded soft-voting guided approach resulted in performance with an average DSC of 0.68 ± 0.12 for the internal test data from developmental dataset, and an average DSC of 0.66 ± 0.18 on the multi-site external data (n = 518), significantly outperforming (p < 0.001) state-of-the-art (SOTA) approaches including nnU-Net and SWIN UNETR. While TTA yielded enhanced performance gains for some of the comparator methods, its impact on our cascaded approach was found to be negligible (DSC: 0.66 ± 0.16). Our approach reliably quantified TMTV, with a correlation of 0.89 with the ground truth (p < 0.001). Furthermore, in terms of visual assessment, concordance between quantitative evaluations and clinician feedback was observed in the majority of cases. The average relative error (ARE) and the absolute error (AE) in TMTV prediction on external multi-centric dataset were ARE = 0.43 ± 0.54 and AE = 157.32 ± 378.12 (mL) for all the external test data (n = 518), and ARE = 0.30 ± 0.22 and AE = 82.05 ± 99.78 (mL) when the 10% outliers (n = 53) were excluded. CONCLUSION: TMTV-Net demonstrates strong performance and generalizability in TMTV segmentation across multi-site external datasets, encompassing various lymphoma subtypes. A negligible reduction of 2% in overall performance during testing on external data highlights robust model generalizability across different centers and cancer types, likely attributable to its training with resampled inputs. Our model is publicly available, allowing easy multi-site evaluation and generalizability analysis on datasets from different institutions.

Saturation Effect of Lipid Accumulation Product (LAP) Index on Spinal Bone Mineral Density: A Population-Based Study.

Lipid accumulation product (LAP) has a positive effect on spinal bone mineral density (BMD). However, once LAP levels exceed 27.26, the rate of spinal BMD increase slow down or even decline. This indicates a biphasic relationship between lipid metabolism and BMD, suggesting potential benefits within a certain range and possible adverse effects beyond that range. This study aimed to investigate the potential association between LAP index and BMD in US adults, as well as to explore the presence of a potential saturation effect in this relationship. This study analyzed data from the National Health and Nutrition Examination Survey (NHANES) spanning from 2007 to 2018. A multiple stepwise regression model was employed to examine the association between LAP index and total spinal BMD. Additionally, a generalized additive model and a smooth curve fitting algorithm were utilized to examine the relationship, and saturation effect study was conducted to determine the saturation level. The calculation formula of LAP used in the study was: (LAP = (waist circumstances (WC) (cm) - 58) × triglyceride (TG) (mmol/L)) for women, and (LAP = (WC (cm) - 65) × TG (mmol/L)) for men. The study involved a total of 7913 participants aged 20 years or older. Through multiple stepwise regression analysis, it was found that individuals with higher LAP scores exhibited higher total spinal BMD. In both the crude and partially adjusted models, total spinal BMD was significantly higher in the highest LAP quartile (Q4) compared to the lowest LAP quartile (Q1) (P < 0.05). Utilizing a generalized additive model and smooth curve, a nonlinear relationship between LAP and total spinal BMD was observed. Furthermore, the study identified the saturation value of LAP to be 27.26, indicating a saturation effect. This research highlights a nonlinear relationship between LAP and total spinal BMD, along with the presence of a saturation effect.

Discovery of novel 5-phenyl-1H-pyrrole-2-carboxylic acids as Keap1-Nrf2 inhibitors for acute lung injury treatment.

Oxidative stress is intricately linked to acute lung injury (ALI) and cerebral ischemic/reperfusion (I/R) injury. The Keap1 (Kelch-like ECH-Associating protein 1)-Nrf2 (nuclear factor erythroid 2-related factor 2)-ARE (antioxidant response element) signaling pathway, recognized as a crucial regulatory mechanism in oxidative stress, holds immense potential for the treatment of both diseases. In our laboratory, we initially screened a compound library and identified compound 3, which exhibited a dissociation constant of 5090 nM for Keap1. To enhance its binding affinity, we developed a novel 5-phenyl-1H-pyrrole-2-carboxylic acid Keap1-Nrf2 inhibitor through scaffold hopping from compound 3. Structure-activity relationship studies identified compound 19 as the most potent, with a KD2 of 42.2 nM against Keap1. Furthermore, compound 19 showed significant protection against LPS-induced injury in BEAS-2B cells and promoted Nrf2 nuclear translocation. Subsequently, we investigated its therapeutic effects in mouse models of ALI injury. Compound 19 effectively alleviated symptoms at doses of 15 mg/kg for ALI injury. Additionally, it facilitated Nrf2 translocation to the nucleus, increased Nrf2 levels, and upregulated the expression of HO-1 and NQO1 in affected tissues.

Successful application of airway pressure release ventilation in a child with severe acute respiratory distress syndrome induced by trauma: a case report.

BACKGROUND: Trauma has been identified as one of the risk factors for acute respiratory distress syndrome. Respiratory support can be further complicated by comorbidities of trauma such as primary or secondary lung injury. Conventional ventilation strategies may not be suitable for all trauma-related acute respiratory distress syndrome. Airway pressure release ventilation has emerged as a potential rescue method for patients with acute respiratory distress syndrome and hypoxemia refractory to conventional mechanical ventilation. However, there is a lack of research on the use of airway pressure release ventilation in children with trauma-related acute respiratory distress syndrome. We report a case of airway pressure release ventilation applied to a child with falling injury, severe acute respiratory distress syndrome, hemorrhagic shock, and bilateral hemopneumothorax. We hope this case report presents a potential option for trauma-related acute respiratory distress syndrome and serves as a basis for future research. CASE PRESENTATION: A 15-year-old female with falling injury who developed severe acute respiratory distress syndrome, hemorrhagic shock, and bilateral hemopneumothorax was admitted to the surgical intensive care unit. She presented refractory hypoxemia despite the treatment of conventional ventilation with deep analgesia, sedation, and muscular relaxation. Lung recruitment was ineffective and prone positioning was contraindicated. Her oxygenation significantly improved after the use of airway pressure release ventilation. She was eventually extubated after 12 days of admission and discharged after 42 days of hospitalization. CONCLUSION: Airway pressure release ventilation may be considered early in the management of trauma patients with severe acute respiratory distress syndrome when prone position ventilation cannot be performed and refractory hypoxemia persists despite conventional ventilation and lung recruitment maneuvers.

A case of septic shock due to delayed diagnosis of Cryptosporidium infection after liver transplantation.

BACKGROUND: Cryptosporidium is recognized as a significant pathogen of diarrhea disease in immunocompromised hosts, and studies have shown that Cryptosporidium infection is high in solid organ transplantation (SOT) patients and often has serious consequences. Because of the lack of specificity of diarrheasymptoms cased by Cryptosporidium infection, it is rarely reported in patients undergoing liver transplantation (LT). It frequently delays diagnosis, coming with severe consequences. In clinical work, diagnosing Cryptosporidium infection in LT patients is also complex but single, and the corresponding anti-infective treatment regimen has not yet been standardized. A rare case of septic shock due to a delayed diagnosis of Cryptosporidium infection after LT and relevant literature are discussed in the passage. CASE PRESENTATION: A patient who had received LT for two years was admitted to the hospital with diarrhea more than 20 days after eating an unclean diet. After failing treatment at a local hospital, he was admitted to Intensive Care Unit after going into septic shock. The patient presented hypovolemia due to diarrhea, which progressed to septic shock. The patient's sepsis shock was controlled after receiving multiple antibiotic combinations and fluid resuscitation. However, the persistent diarrhea, as the culprit of the patient's electrolyte disturbance, hypovolemia, and malnutrition, was unsolved. The causative agent of diarrhea, Cryptosporidium infection, was identified by colonoscopy, faecal antacid staining, and blood high-throughput sequencing (NGS). The patient was treated by reducing immunosuppression and Nitazoxanide (NTZ), which proved effective in this case. CONCLUSION: When LT patients present with diarrhea, clinicians should consider the possibility of Cryptosporidium infection, in addition to screening for conventional pathogens. Tests such as colonoscopy, stool antacid staining and blood NGS sequencing can help diagnose and treat of Cryptosporidium infection early and avoid serious consequences of delayed diagnosis. In treating Cryptosporidium infection in LT patients, the focus should be on the patient's immunosuppressive therapy, striking a balance between anti-immunorejection and anti-infection should be sought. Based on practical experience, NTZ therapy in combination with controlled CD4 + T cells at 100-300/mm3 was highly effective against Cryptosporidium without inducing immunorejection.

Diagnosis of Coronavirus Disease 2019 Pneumonia by Using Chest Radiography: Value of Artificial Intelligence.

Background Radiologists are proficient in differentiating between chest radiographs with and without symptoms of pneumonia but have found it more challenging to differentiate coronavirus disease 2019 (COVID-19) pneumonia from non-COVID-19 pneumonia on chest radiographs. Purpose To develop an artificial intelligence algorithm to differentiate COVID-19 pneumonia from other causes of abnormalities at chest radiography. Materials and Methods In this retrospective study, a deep neural network, CV19-Net, was trained, validated, and tested on chest radiographs in patients with and without COVID-19 pneumonia. For the chest radiographs positive for COVID-19, patients with reverse transcription polymerase chain reaction results positive for severe acute respiratory syndrome coronavirus 2 with findings positive for pneumonia between February 1, 2020, and May 30, 2020, were included. For the non-COVID-19 chest radiographs, patients with pneumonia who underwent chest radiography between October 1, 2019, and December 31, 2019, were included. Area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were calculated to characterize diagnostic performance. To benchmark the performance of CV19-Net, a randomly sampled test data set composed of 500 chest radiographs in 500 patients was evaluated by the CV19-Net and three experienced thoracic radiologists. Results A total of 2060 patients (5806 chest radiographs; mean age, 62 years ± 16 [standard deviation]; 1059 men) with COVID-19 pneumonia and 3148 patients (5300 chest radiographs; mean age, 64 years ± 18; 1578 men) with non-COVID-19 pneumonia were included and split into training and validation and test data sets. For the test set, CV19-Net achieved an AUC of 0.92 (95% CI: 0.91, 0.93). This corresponded to a sensitivity of 88% (95% CI: 87, 89) and a specificity of 79% (95% CI: 77, 80) by using a high-sensitivity operating threshold, or a sensitivity of 78% (95% CI: 77, 79) and a specificity of 89% (95% CI: 88, 90) by using a high-specificity operating threshold. For the 500 sampled chest radiographs, CV19-Net achieved an AUC of 0.94 (95% CI: 0.93, 0.96) compared with an AUC of 0.85 (95% CI: 0.81, 0.88) achieved by radiologists. Conclusion CV19-Net was able to differentiate coronavirus disease 2019-related pneumonia from other types of pneumonia, with performance exceeding that of experienced thoracic radiologists. © RSNA, 2021 Online supplemental material is available for this article.

Pterostilbene and its nicotinate derivative ameliorated vascular endothelial senescence and elicited endothelium-dependent relaxations via activation of sirtuin 1.

Vascular endothelial cell senescence is a leading cause of age-associated diseases and cardiovascular diseases. Interventions and therapies targeting endothelial cell senescence and dysfunction would have important clinical implications. This study evaluated the effect of 10 resveratrol analogues, including pterostilbene (Pts) and its derivatives, against endothelial senescence and dysfunction. All the tested compounds at the concentrations from 10-9 M to 10-6 M did not show cytotoxicity in endothelial cells by MTT assay. Among the 10 resveratrol analogues, Pts and Pts nicotinate attenuated the expression of senescence-associated ß-galactosidase, downregulated p21 and p53, and increased the production of nitric oxide (NO) in both angiotensin II - and hydrogen peroxide - induced endothelial senescence models. In addition, Pts and Pts nicotinate elicited endothelium-dependent relaxations, which were attenuated in the presence of endothelial NO synthase (eNOS) inhibitor L-NAME or sirtuin 1 (SIRT1) inhibitor sirtinol. Pts and Pts nicotinate did not alter SIRT1 expression but enhanced its activity. Both Pts and Pts nicotinate have high binding activities with SIRT1, according to surface plasmon resonance results and the molecular docking analysis. Inhibition of SIRT1 by sirtinol reversed the anti-senescent effects of Pts and Pts nicotinate. Moreover, Pts and Pts nicotinate shared similar ADME (absorption, distribution, metabolism, excretion) profiles and physiochemical properties. This study suggests that the Pts and Pts nicotinate ameliorate vascular endothelial senescence and elicit endothelium-dependent relaxations via activation of SIRT1. These two compounds may be potential drugs for the treatment of cardiovascular diseases related to endothelial senescence and dysfunction.

Thin-layer chromatographic identification of Chinese propolis using chemometric fingerprinting.

INTRODUCTION: Poplar tree gum has a similar chemical composition and appearance to Chinese propolis (bee glue) and has been widely used as a counterfeit propolis because Chinese propolis is typically the poplar-type propolis, the chemical composition of which is determined mainly by the resin of poplar trees. The discrimination of Chinese propolis from poplar tree gum is a challenging task. OBJECTIVE: To develop a rapid thin-layer chromatographic (TLC) identification method using chemometric fingerprinting to discriminate Chinese propolis from poplar tree gum. METHODS: A new TLC method using a combination of ammonia and hydrogen peroxide vapours as the visualisation reagent was developed to characterise the chemical profile of Chinese propolis. Three separate people performed TLC on eight Chinese propolis samples and three poplar tree gum samples of varying origins. Five chemometric methods, including similarity analysis, hierarchical clustering, k-means clustering, neural network and support vector machine, were compared for use in classifying the samples based on their densitograms obtained from the TLC chromatograms via image analysis. RESULTS: Hierarchical clustering, neural network and support vector machine analyses achieved a correct classification rate of 100% in classifying the samples. A strategy for TLC identification of Chinese propolis using chemometric fingerprinting was proposed and it provided accurate sample classification. CONCLUSION: The study has shown that the TLC identification method using chemometric fingerprinting is a rapid, low-cost method for the discrimination of Chinese propolis from poplar tree gum and may be used for the quality control of Chinese propolis.

ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax.

Radiology narrative reports often describe characteristics of a patient's disease, including its location, size, and shape. Motivated by the recent success of multimodal learning, we hypothesized that this descriptive text could guide medical image analysis algorithms. We proposed a novel vision-language model, ConTEXTual Net, for the task of pneumothorax segmentation on chest radiographs. ConTEXTual Net extracts language features from physician-generated free-form radiology reports using a pre-trained language model. We then introduced cross-attention between the language features and the intermediate embeddings of an encoder-decoder convolutional neural network to enable language guidance for image analysis. ConTEXTual Net was trained on the CANDID-PTX dataset consisting of 3196 positive cases of pneumothorax with segmentation annotations from 6 different physicians as well as clinical radiology reports. Using cross-validation, ConTEXTual Net achieved a Dice score of 0.716±0.016, which was similar to the degree of inter-reader variability (0.712±0.044) computed on a subset of the data. It outperformed vision-only models (Swin UNETR: 0.670±0.015, ResNet50 U-Net: 0.677±0.015, GLoRIA: 0.686±0.014, and nnUNet 0.694±0.016) and a competing vision-language model (LAVT: 0.706±0.009). Ablation studies confirmed that it was the text information that led to the performance gains. Additionally, we show that certain augmentation methods degraded ConTEXTual Net's segmentation performance by breaking the image-text concordance. We also evaluated the effects of using different language models and activation functions in the cross-attention module, highlighting the efficacy of our chosen architectural design.

Genetic associations between autoimmune diseases and the risks of severe sepsis and 28-day mortality: a two-sample Mendelian randomization study.

Background: Autoimmune diseases exhibit heterogenous dysregulation of pro-inflammatory or anti-inflammatory cytokine expression, akin to the pathophysiology of sepsis. It is speculated that individuals with autoimmune diseases may have an increased likelihood of developing sepsis and face elevated mortality risks following septic events. However, current observational studies have not yielded consistent conclusions. This study aims to explore the causal relationship between autoimmune diseases and the risks of sepsis and mortality using Mendelian randomization (MR) analysis. Methods: We conducted a two-sample MR study involving a European population, with 30 autoimmune diseases as the exposure factors. To assess causal relationships, we employed the inverse variance-weighted (IVW) method and used Cochran's Q test for heterogeneity, as well as the MR pleiotropy residual sum and outlier (MR-PRESSO) global test for potential horizontal pleiotropy. Results: Genetically predicted Crohn's disease (ß = 0.067, se = 0.034, p = 0.046, OR = 1.069, 95% CI = 1.001-1.141) and idiopathic thrombocytopenic (ß = 0.069, se = 0.031, p = 0.023, OR = 1.071, 95% CI = 1.009-1.136) were positively associated with an increased risk of sepsis in critical care. Conversely, rheumatoid arthritis (ß = -0.104, se = 0.047, p = 0.025, OR = 0.901, 95% CI = 0.823-0.987), ulcerative colitis (ß = -0.208, se = 0.084, p = 0.013, OR = 0.812, 95% CI = 0.690-0.957), and narcolepsy (ß = -0.202, se = 0.092, p = 0.028, OR = 0.818, 95% CI = 0.684-0.978) were associated with a reduced risk of sepsis in critical care. Moreover, Crohn's disease (ß = 0.234, se = 0.067, p = 0.001, OR = 1.263, 95% CI = 1.108-1.440) and idiopathic thrombocytopenic (ß = 0.158, se = 0.061, p = 0.009, OR = 1.171, 95% CI = 1.041-1.317) were also linked to an increased risk of 28-day mortality of sepsis in critical care. In contrast, multiple sclerosis (ß = -0.261, se = 0.112, p = 0.020, OR = 0.771, 95% CI = 0.619-0.960) and narcolepsy (ß = -0.536, se = 0.184, p = 0.003, OR = 0.585, 95% CI = 0.408-0.838) were linked to a decreased risk of 28-day mortality of sepsis in critical care. Conclusion: This MR study identified causal associations between certain autoimmune diseases and risks of sepsis in critical care, and 28-day mortality in the European population. These findings suggest that exploring the mechanisms underlying autoimmune diseases may offer new diagnostic and therapeutic strategies for sepsis prevention and treatment.

Associations between serum albumin level trajectories and clinical outcomes in sepsis patients in ICU: insights from longitudinal group trajectory modeling.

Background: Sepsis triggers a strong inflammatory response, often leading to organ failure and high mortality. The role of serum albumin levels in sepsis is critical but not fully understood, particularly regarding the significance of albumin level changes over time. This study utilized Group-based Trajectory Modeling (GBTM) to investigate the patterns of serum albumin changes and their impact on sepsis outcomes. Methods: We conducted a retrospective analysis on ICU patients from West China Hospital (2015-2022), employing GBTM to study serum albumin fluctuations within the first week of ICU admission. The study factored in demographics, clinical parameters, and comorbidities, handling missing data through multiple imputation. Outcomes assessed included 28-day mortality, overall hospital mortality, and secondary complications such as AKI and the need for mechanical ventilation. Results: Data from 1,950 patients revealed four serum albumin trajectories, showing distinct patterns of consistently low, increasing, moderate, and consistently high levels. These groups differed significantly in mortality, with the consistently low level group experiencing the highest mortality. No significant difference in 28-day mortality was observed among the other groups. Subgroup analysis did not alter these findings. Conclusion: The study identified four albumin trajectory groups in sepsis patients, highlighting that those with persistently low levels had the worst outcomes, while those with increasing levels had the best. Stable high levels above 30 g/L did not change outcomes significantly. These findings can inform clinical decisions, helping to identify high-risk patients early and tailor treatment approaches.

Personalized Impression Generation for PET Reports Using Large Language Models.

Large language models (LLMs) have shown promise in accelerating radiology reporting by summarizing clinical findings into impressions. However, automatic impression generation for whole-body PET reports presents unique challenges and has received little attention. Our study aimed to evaluate whether LLMs can create clinically useful impressions for PET reporting. To this end, we fine-tuned twelve open-source language models on a corpus of 37,370 retrospective PET reports collected from our institution. All models were trained using the teacher-forcing algorithm, with the report findings and patient information as input and the original clinical impressions as reference. An extra input token encoded the reading physician's identity, allowing models to learn physician-specific reporting styles. To compare the performances of different models, we computed various automatic evaluation metrics and benchmarked them against physician preferences, ultimately selecting PEGASUS as the top LLM. To evaluate its clinical utility, three nuclear medicine physicians assessed the PEGASUS-generated impressions and original clinical impressions across 6 quality dimensions (3-point scales) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. When physicians assessed LLM impressions generated in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08/5. On average, physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P = 0.41). In summary, our study demonstrated that personalized impressions generated by PEGASUS were clinically useful in most cases, highlighting its potential to expedite PET reporting by automatically drafting impressions.

Automatic Quantification of Serial PET/CT Images for Pediatric Hodgkin Lymphoma Patients Using a Longitudinally-Aware Segmentation Network.

Purpose: Automatic quantification of longitudinal changes in PET scans for lymphoma patients has proven challenging, as residual disease in interim-therapy scans is often subtle and difficult to detect. Our goal was to develop a longitudinally-aware segmentation network (LAS-Net) that can quantify serial PET/CT images for pediatric Hodgkin lymphoma patients. Materials and Methods: This retrospective study included baseline (PET1) and interim (PET2) PET/CT images from 297 patients enrolled in two Children's Oncology Group clinical trials (AHOD1331 and AHOD0831). LAS-Net incorporates longitudinal cross-attention, allowing relevant features from PET1 to inform the analysis of PET2. Model performance was evaluated using Dice coefficients for PET1 and detection F1 scores for PET2. Additionally, we extracted and compared quantitative PET metrics, including metabolic tumor volume (MTV) and total lesion glycolysis (TLG) in PET1, as well as qPET and ΔSUVmax in PET2, against physician measurements. We quantified their agreement using Spearman's ρ correlations and employed bootstrap resampling for statistical analysis. Results: LAS-Net detected residual lymphoma in PET2 with an F1 score of 0.606 (precision/recall: 0.615/0.600), outperforming all comparator methods (P<0.01). For baseline segmentation, LAS-Net achieved a mean Dice score of 0.772. In PET quantification, LAS-Net's measurements of qPET, ΔSUVmax, MTV and TLG were strongly correlated with physician measurements, with Spearman's ρ of 0.78, 0.80, 0.93 and 0.96, respectively. The quantification performance remained high, with a slight decrease, in an external testing cohort. Conclusion: LAS-Net demonstrated significant improvements in quantifying PET metrics across serial scans, highlighting the value of longitudinal awareness in evaluating multi-time-point imaging datasets.

Prediction of upcoming urinary tract infection after intracerebral hemorrhage: a machine learning approach based on statistics collected at multiple time points.

Purpose: Accurate prediction of urinary tract infection (UTI) following intracerebral hemorrhage (ICH) can significantly facilitate both timely medical interventions and therapeutic decisions in neurocritical care. Our study aimed to propose a machine learning method to predict an upcoming UTI by using multi-time-point statistics. Methods: A total of 110 patients were identified from a neuro-intensive care unit in this research. Laboratory test results at two time points were chosen: Lab 1 collected at the time of admission and Lab 2 collected at the time of 48 h after admission. Univariate analysis was performed to investigate if there were statistical differences between the UTI group and the non-UTI group. Machine learning models were built with various combinations of selected features and evaluated with accuracy (ACC), sensitivity, specificity, and area under the curve (AUC) values. Results: Corticosteroid usage (p < 0.001) and daily urinary volume (p < 0.001) were statistically significant risk factors for UTI. Moreover, there were statistical differences in laboratory test results between the UTI group and the non-UTI group at the two time points, as suggested by the univariate analysis. Among the machine learning models, the one incorporating clinical information and the rate of change in laboratory parameters outperformed the others. This model achieved ACC = 0.773, sensitivity = 0.785, specificity = 0.762, and AUC = 0.868 during training and 0.682, 0.685, 0.673, and 0.751 in the model test, respectively. Conclusion: The combination of clinical information and multi-time-point laboratory data can effectively predict upcoming UTIs after ICH in neurocritical care.

A quality-checked and physics-constrained deep learning method to estimate material basis images from single-kV contrast-enhanced chest CT scans.

BACKGROUND: Single-kV CT imaging is one of the primary imaging methods in radiology practices. However, it does not provide material basis images for some subtle lesion characterization tasks in clinical diagnosis. PURPOSE: To develop a quality-checked and physics-constrained deep learning (DL) method to estimate material basis images from single-kV CT data without resorting to dual-energy CT acquisition schemes. METHODS: Single-kV CT images are decomposed into two material basis images using a deep neural network. The role of this network is to generate a feature space with 64 template features with the same matrix dimensions of the input single-kV CT image. These 64 template image features are then combined to generate the desired material basis images with different sets of combination coefficients, one for each material basis image. Dual-energy CT image acquisitions with two separate kVs were curated to generate paired training data between a single-kV CT image and the corresponding two material basis images. To ensure the obtained two material basis images are consistent with the encoded spectral information in the actual projection data, two physics constraints, that is, (1) effective energy of each measured projection datum that characterizes the beam hardening in data acquisitions and (2) physical factors of scanners such as detector and tube characteristics, are incorporated into the end-to-end training. The entire architecture is referred to as Deep-En-Chroma in this paper. In the application stage, the generated material basis images are sent to a deep quality check (Deep-QC) network to assess the quality of estimated images and to report the pixel-wise estimation errors for users. The models were developed using 5592 training and validation pairs generated from 48 clinical cases. Additional 1526 CT images from another 13 patients were used to evaluate the quantitative accuracy of water and iodine basis images estimated by Deep-En-Chroma. RESULTS: For the iodine basis images estimated by Deep-En-Chroma, the mean difference with respect to dual-energy CT is -0.25 mg/mL, and the agreement limits are [-0.75 mg/mL, +0.24 mg/mL]. For the water basis images estimated by Deep-En-Chroma, the mean difference with respect to dual-energy CT is 0.0 g/mL, and the agreement limits are [-0.01 g/mL, 0.01 g/mL]. Across the test cohort, the median [25th, 75th percentiles] root mean square errors between the Deep-En-Chroma and dual-energy material images are 14 [12, 16] mg/mL for the water images and 0.73 [0.64, 0.80] mg/mL for the iodine images. When significant errors are present in the estimated material basis images, Deep-QC can capture these errors and provide pixel-wise error maps to inform users whether the DL results are trustworthy. CONCLUSIONS: The Deep-En-Chroma network provides a new pathway to estimating the clinically relevant material basis images from single-kV CT data and the Deep-QC module to inform end-users of the accuracy of the DL material basis images in practice.

Automatic Personalized Impression Generation for PET Reports Using Large Language Models.

Purpose: To determine if fine-tuned large language models (LLMs) can generate accurate, personalized impressions for whole-body PET reports. Materials and Methods: Twelve language models were trained on a corpus of PET reports using the teacher-forcing algorithm, with the report findings as input and the clinical impressions as reference. An extra input token encodes the reading physician's identity, allowing models to learn physician-specific reporting styles. Our corpus comprised 37,370 retrospective PET reports collected from our institution between 2010 and 2022. To identify the best LLM, 30 evaluation metrics were benchmarked against quality scores from two nuclear medicine (NM) physicians, with the most aligned metrics selecting the model for expert evaluation. In a subset of data, model-generated impressions and original clinical impressions were assessed by three NM physicians according to 6 quality dimensions (3-point scale) and an overall utility score (5-point scale). Each physician reviewed 12 of their own reports and 12 reports from other physicians. Bootstrap resampling was used for statistical analysis. Results: Of all evaluation metrics, domain-adapted BARTScore and PEGASUSScore showed the highest Spearman's ρ correlations (ρ=0.568 and 0.563) with physician preferences. Based on these metrics, the fine-tuned PEGASUS model was selected as the top LLM. When physicians reviewed PEGASUS-generated impressions in their own style, 89% were considered clinically acceptable, with a mean utility score of 4.08 out of 5. Physicians rated these personalized impressions as comparable in overall utility to the impressions dictated by other physicians (4.03, P=0.41). Conclusion: Personalized impressions generated by PEGASUS were clinically useful, highlighting its potential to expedite PET reporting.

Current Status and Research Trends of Panax Between 1900-2019: A Bibliometric Analysis.

OBJECTIVE: To investigate the current status and further development of Panax genus and 6 important individual species including P. notoginseng, P. quinquefolium, P. vietnamensis, P. japonicus, P. stipuleanatus and P. zingiberensis. METHODS: The bibliometric analysis was based on the Web of Science core database platform from Thomson Reuters. Totally, 7,574 records of scientific research of Panax species published from 1900-2019 were analyzed. The statistical and visualization analysis was performed by CiteSpace and HistCite software. RESULTS: The academic research of Panax species increase promptly. Plant science is the main research field while research and experimental medicine and agricultural engineering will be the further development tendency. Particularly, the discrimination research of P. notoginseng will be the research tendency among Panax species, especially diversity research. In addition, P. vietnamensis deserves more attention in the genus Panax. CONCLUSION: This research provides a reference for further research of the genus and individual species.

Charge transfer dynamics of 3,4,9,10-perylene-tetracarboxylic-dianhydride molecules on Au(111) probed by resonant photoemission spectroscopy.

Charge transfer dynamics across the lying-down 3,4,9,10-perylene-tetracarboxylic-dianhydride (PTCDA) organic semiconductor molecules on Au(111) interface has been investigated using the core-hole clock implementation of resonant photoemission spectroscopy. It is found that the charge transfer time scale at the PTCDA∕Au(111) interface is much larger than the C 1s core-hole lifetime of 6 fs, indicating weak electronic coupling between PTCDA and the gold substrate due to the absence of chemical reaction and∕or bonding.

Pseudo-CT generation from multi-parametric MRI using a novel multi-channel multi-path conditional generative adversarial network for nasopharyngeal carcinoma patients.

PURPOSE: To develop and evaluate a novel method for pseudo-CT generation from multi-parametric MR images using multi-channel multi-path generative adversarial network (MCMP-GAN). METHODS: Pre- and post-contrast T1-weighted (T1-w), T2-weighted (T2-w) MRI, and treatment planning CT images of 32 nasopharyngeal carcinoma (NPC) patients were employed to train a pixel-to-pixel MCMP-GAN. The network was developed based on a 5-level Residual U-Net (ResU-Net) with the channel-based independent feature extraction network to generate pseudo-CT images from multi-parametric MR images. The discriminator with five convolutional layers was added to distinguish between the real CT and pseudo-CT images, improving the nonlinearity and prediction accuracy of the model. Eightfold cross validation was implemented to validate the proposed MCMP-GAN. The pseudo-CT images were evaluated against the corresponding planning CT images based on mean absolute error (MAE), peak signal-to-noise ratio (PSNR), Dice similarity coefficient (DSC), and Structural similarity index (SSIM). Similar comparisons were also performed against the multi-channel single-path GAN (MCSP-GAN), the single-channel single-path GAN (SCSP-GAN). RESULTS: It took approximately 20 h to train the MCMP-GAN model on a Quadro P6000, and less than 10 s to generate all pseudo-CT images for the subjects in the test set. The average head MAE between pseudo-CT and planning CT was 75.7 ± 14.6 Hounsfield Units (HU) for MCMP-GAN, significantly (P-values < 0.05) lower than that for MCSP-GAN (79.2 ± 13.0 HU) and SCSP-GAN (85.8 ± 14.3 HU). For bone only, the MCMP-GAN yielded a smaller mean MAE (194.6 ± 38.9 HU) than MCSP-GAN (203.7 ± 33.1 HU), SCSP-GAN (227.0 ± 36.7 HU). The average PSNR of MCMP-GAN (29.1 ± 1.6) was found to be higher than that of MCSP-GAN (28.8 ± 1.2) and SCSP-GAN (28.2 ± 1.3). In terms of metrics for image similarity, MCMP-GAN achieved the highest SSIM (0.92 ± 0.02) but did not show significantly improved bone DSC results in comparison with MCSP-GAN. CONCLUSIONS: We developed a novel multi-channel GAN approach for generating pseudo-CT from multi-parametric MR images. Our preliminary results in NPC patients showed that the MCMP-GAN method performed apparently superior to the U-Net-GAN and SCSP-GAN, and slightly better than MCSP-GAN.