Baseline CT Scans Research Articles

Development of a nomogram-based model incorporating radiomic features from follow-up longitudinal lung CT images to distinguish invasive adenocarcinoma from benign lesions: a retrospective study

PurposeTo develop and validate a radiomic model for differentiating pulmonary invasive adenocarcinomas from benign lesions based on follow-up longitudinal CT images.MethodsThis is a retrospective study including 336 patients (161 with invasive adenocarcinomas and 175 with benign lesions) who underwent baseline (T0) and follow-up (T1) CT scans from January 2016 to June 2022. The patients were randomized in a 7:3 ratio into training and test sets. Radiomic features were extracted from lesion volumes of interest on longitudinal CT images at T0 and T1. Differences in radiomic features between T1 and T0 were defined as delta-radiomic features. Logistic regression was used to build models based on clinicoradiological (CR), T0, T1, and delta radiomic features and compute signatures. Finally, a nomogram based on the CR, T0, T1 and delta signatures was constructed. Model performance was evaluated for calibration, discrimination, and clinical utility.ResultsThe T1 radiomic model was superior to the other independent models. In the training set, it had an area under the curve (AUC) of 0.858), superior to the CR model (AUC 0.694), the T0 radiomic model (AUC 0.825), and the delta radiomic model (AUC 0.734). In the test set, it had an AUC of 0.817, again outperforming the CR model (AUC 0.578), the T0 radiomic model (AUC 0.789), and the delta radiomic model (AUC 0.647). The nomogram incorporating the CR, T0, T1 and delta signatures showed the best predictive performance in both the training (AUC: 0.906) and test sets (AUC: 0.856), and it exhibited excellent fit with calibration curves. Decision curve analysis provided additional validation of the clinical utility of the nomogram.ConclusionA nomogram utilizing radiomic features extracted from longitudinal CT images can enhance the discriminative capability between pulmonary invasive adenocarcinomas and benign lesions.

Prognostic Value of Sarcopenia in Elderly Patients with Metastatic Non-Small-Cell Lung Cancer Undergoing Radiotherapy

Background: Sarcopenia, a syndrome characterized by age-related loss of muscle mass and function, lacks universally accepted diagnostic criteria, particularly for its role as a prognostic factor in elderly patients with non-small-cell lung cancer (NSCLC). This study aimed to evaluate the prognostic significance of sarcopenia, assessed by psoas muscle size on baseline CT scans, in patients over 70 years of age with metastatic NSCLC. Methods: We retrospectively analyzed 85 elderly patients undergoing palliative radiation therapy between August 2022 and July 2024. Using morphometric analysis of psoas size, we investigated its correlation with overall survival (OS) and progression-free survival (PFS). Results: Our results showed that decreased psoas size was significantly associated with shorter OS and PFS, with median OS of 10 months and PFS of 4 months in sarcopenic patients compared to longer survival times in non-sarcopenic patients. Median survival of non-sarcopenic vs. sarcopenic patients was 21 ± 7 months (muscle area > median) versus 5 ± 2.3 months (muscle area < median). Multivariate analysis confirmed that psoas size, along with ECOG performance status and treatment of primary NSCLC, was a significant predictor of survival. Discussion: These findings suggest that psoas muscle size is a valuable prognostic marker for elderly NSCLC patients, potentially guiding treatment decisions and patient management. Further research is needed to validate these results and refine prognostic models for this population.

A Study in Pituitary Neuroendocrine Tumors (PitNETs): Real-Life Data Amid Baseline and Serial CT Scans.

We aimed to evaluate CT-related PIs diameters following 3 aspects: a cross-sectional analysis based on the age' groups at first PI diagnosis and on the gender distribution and a longitudinal analysis in PIs with <0.5 cm versus ≥0.5 cm as the largest diameter at baseline. A retrospective, real-life, multi-centric study in adults with NF micro-PIs was performed. 208 subjects (92.79% females) were included (average age 43.18 ± 12.58 y). The mean largest diameter (between transversal and longitudinal diameters) was 0.55 ± 0.16cm, and 44.71% of the PIs were located on the right part of the pituitary gland. The patients were divided into 10 age-groups (21 to 70 y), and no difference was observed between the mean largest diameters of these groups (p = 0.693). Females and males had a similar age at PIs diagnosis (p = 0.353), transverse diameter (p = 0.910), longitudinal diameter (p = 0.229), and PI location (p = 0.368). Serial CT scans (2 to 4 per patient) included a median (Q1-Q3) of 20 (12-36) months for the second CT scan, 58 (36-84) for the third CT scan, 78 (53-118) for the fourth CT scan, and a total follow-up between baseline and last CT scan of 40 (13-72) months. The initial largest diameter correlated with the diameter change between the baseline and the last CT (r = -0.575, p = 0.000). The largest PI diameter was similar between each serial CT scan (p = 0.840). According to the cut-off value of 0.5 cm diameter (for initial largest tumor diameter), group A (N = 78, 37.50%, <0.5 cm) and group B (N = 130, 62. 50%; ≥0.5 cm) had similar age at baseline (43.83 ± 12.72 versus 42.79 ± 12.54 y, p = 0.565) and PIs locations (right, left, median). The largest PI diameter remained smaller in group A versus B amid the second CT scan (p = 0.000) and the last CT scan (p = 0.009). The largest diameter change from the first to the last CT scan showed an increase in group A (median of +0.10 cm, p = 0.000) and a decrease in group B (median of -0.01 cm, p = 0.002); this diameter change was different in group A versus B (p = 0.000). The cumulative probability of tumor-growth-free survival showed different survival functions (log rank p = 0.000): group A exhibited a more gradual decrease versus B; at 60 months, the cumulative probability was 0.32 ± 0.08 for group A versus 0.75 ± 0.07 for B. During follow-up, all PIs remained NF, and no hypopituitarism was noted (as limits, we mention that dynamic hypopituitarism testing was selectively performed). NF micro-PIs in adults showed a similar age and sex distribution. During follow-up, PIs with a largest diameter < 0.5 cm increased after a median of 40 months but remained <1 cm, while in PIs with ≥0.5 cm, the largest diameter decreased. This highlights a lower predictability in tumor behavior than expected, particularly in larger micro-PIs that, overall, remained without relevant clinical implications after surveillance.

Appendiceal adenocarcinoma-patterns of tumor spread and prognosis

IntroductionAppendiceal adenocarcinoma represents a diagnostic and therapeutic challenge since it is prone to early lymphatic and peritoneal spread. We aimed to analyze the proportion of lymph node metastases in completion right hemicolectomy specimens, risk factors for peritoneal metastases (PM), and prognosis after definitive treatment. MethodsNinety-three patients with appendiceal adenocarcinoma scheduled for cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS + HIPEC) in Uppsala 2004–2020 were identified from a prospectively maintained registry. Risk factors for PM were assessed based on the presence (CT + group, n = 55) or absence (CT – group, n = 37) of visible PM at baseline CT scan. Prognostic factors were analyzed based on the actual presence (PM group, n = 66) or absence (no PM group, n = 27) of PM. ResultsThe median age was 60 (26–78). Forty-eight patients were women. Resection of PM at initial surgery indicated an 80 % risk of finding PM at a follow-up exploration. R1 appendectomy and perforated appendix had a similar risk for PM (24 %,26 %) which increased to 38 % if both were present.Regional lymph node metastases occurred in 31 % in the CT + group vs. 14 % in the CT − group (p = 0.005) and was associated with poor survival HR 5.16 (1.49–17.81).The 5-year OS and DFS rates were 54 % and 29 % in the PM group. ConclusionsPatients with certain risk factors have a high likelihood of PM despite a normal CT scan, which justifies selective exploration at a HIPEC center. Regional lymph node spread supports the current practice of completion right hemicolectomy and is a significant prognostic factor.

U-Net-based computed tomography quantification of viral pneumonia can predict fibrotic interstitial lung abnormalities at 3-month follow-up.

Given the high prevalence of fibrotic interstitial lung abnormalities (ILAs) post-COVID-19, this study aims to evaluate the effectiveness of quantitative CT features in predicting fibrotic ILAs at 3-month follow-up. This retrospective study utilized cohorts from distinct clinical settings: the training dataset comprised individuals presenting at the fever clinic and emergency department, while the validation dataset included patients hospitalized with COVID-19 pneumonia. They were classified into fibrotic group and nonfibrotic group based on whether the fibrotic ILAs were present at follow-up. A U-Net-based AI tool was used for quantification of both pneumonia lesions and pulmonary blood volumes. Receiver operating characteristic (ROC) curve analysis and multivariate analysis were used to assess their predictive abilities for fibrotic ILAs. Among the training dataset, 122 patients (mean age of 68 years ±16 [standard deviation], 73 men), 55.74% showed fibrotic ILAs at 3-month follow-up. The multivariate analysis identified the pneumonia volume [PV, odd ratio (OR) 3.28, 95% confidence interval (CI): 1.20-9.31, p = 0.02], consolidation volume (CV, OR 3.77, 95% CI: 1.37-10.75, p = 0.01), ground-glass opacity volume (GV, OR 3.38, 95% CI: 1.26-9.38, p = 0.02), pneumonia mass (PM, OR 3.58, 95% CI: 1.28-10.46, p = 0.02), and the CT score (OR 12.06, 95% CI: 3.15-58.89, p < 0.001) as independent predictors of fibrotic ILAs, and all quantitative parameters were as effective as CT score (all p > 0.05). And the area under the curve (AUC) values were PV (0.79), GV (0.78), PM (0.79), CV (0.80), and the CT score (0.77). The validation dataset, comprising 45 patients (mean age 67.29 ± 14.29 years, 25 males) with 57.78% showing fibrotic ILAs at follow-up, confirmed the predictive validity of these parameters with AUC values for PV (0.86), CV (0.90), GV (0.83), PM (0.88), and the CT score (0.85). Additionally, the percentage of blood volume in vessels <5mm2 relative to the total pulmonary blood volume (BV5%) was significantly lower in patients with fibrotic ILAs (p = 0.048) compared to those without. U-Net based quantification of pneumonia lesion and BV5% on baseline CT scan has the potential to predict fibrotic ILAs at follow-up in COVID-19 patients.

Target-based deep learning network surveillance of non-contrast computed tomography for small infarct core of acute ischemic stroke.

Rapid diagnosis of acute ischemic stroke (AIS) is critical to achieve positive outcomes and prognosis. This study aimed to construct a model to automatically identify the infarct core based on non-contrast-enhanced CT images, especially for small infarcts. The baseline CT scans of AIS patients, who had DWI scans obtained within less than 2 h apart, were included in this retrospective study. A modified Target-based deep learning model of YOLOv5 was developed to detect infarctions on CT. Randomly selected CT images were used for testing and evaluated by neuroradiologists and the model, using the DWI as a reference standard. Intraclass correlation coefficient (ICC) and weighted kappa were calculated to assess the agreement. The paired chi-square test was used to compare the diagnostic efficacy of physician groups and automated models in subregions. p < 0.05 was considered statistically significant. Five hundred and eighty four AIS patients were enrolled in total, finally 275 cases were eligible. Modified YOLOv5 perform better with increased precision (0.82), recall (0.81) and mean average precision (0.79) than original YOLOv5. Model showed higher consistency to the DWI-ASPECTS scores (ICC = 0.669, κ = 0.447) than neuroradiologists (ICC = 0.452, κ = 0.247). The sensitivity (75.86% vs. 63.79%), specificity (98.87% vs. 95.02%), and accuracy (96.20% vs. 91.40%) were better than neuroradiologists. Automatic model had better diagnostic efficacy than physician diagnosis in the M6 region (p = 0.039). The deep learning model was able to detect small infarct core on CT images more accurately. It provided the infarct portion and extent, which is valuable in assessing the severity of disease and guiding treatment procedures.

Transfer learning approach in pre-treatment CT images to predict therapeutic response in advanced malignant pleural mesothelioma.

Malignant pleural mesothelioma (MPM) is a poor-prognosis disease. Owing to the recent availability of new therapeutic options, there is a need to better assess prognosis. The initial clinical response could represent a useful parameter. We proposed a transfer learning approach to predict an initial treatment response starting from baseline CT scans of patients with advanced/unresectable MPM undergoing first-line systemic therapy. The therapeutic response has been assessed according to the mRECIST criteria by CT scan at baseline and after two to three treatment cycles. We used three slices of baseline CT scan as input to the pre-trained convolutional neural network as a radiomic feature extractor. We identified a feature subset through a double feature selection procedure to train a binary SVM classifier to discriminate responders (partial response) from non-responders (stable or disease progression). The performance of the prediction classifiers was evaluated with an 80:20 hold-out validation scheme. We have evaluated how the developed model was robust to variations in the slices selected by the radiologist. In our dataset, 25 patients showed an initial partial response, whereas 13 patients showed progressive or stable disease. On the independent test, the proposed model achieved a median AUC and accuracy of 86.67% and 87.50%, respectively. The proposed model has shown high performance even by varying the reference slices. Novel tools could help to improve the prognostic assessment of patients with MPM and to better identify subgroups of patients with different therapeutic responsiveness.

Temporalis muscle thickness as a predictor of functional outcome after reperfusion therapies for acute ischemic stroke: a retrospective, cohort study.

Sarcopenia, defined as the loss of skeletal muscle mass, has been associated with a worse functional outcome after stroke. Measurement of temporal muscle thickness (TMT) has been introduced as an easily obtainable surrogate marker to identify patients with sarcopenia. Our study aims to investigate the correlation between pre-stroke sarcopenia, measured by TMT assessment, and functional outcome in patients treated with revascularization procedures for acute ischemic stroke. We included consecutive adult patients who underwent thrombolysis, endovascular thrombectomy or both for acute ischemic stroke at our Centre from January 2020 to June 2022. Besides collecting baseline clinical and neuroradiological features, TMT was measured on brain computed tomography scans according to a standardized protocol. Modified Rankin Scale (mRS) scores at 3months represented the main endpoint of functional outcome. A total of 261 patients were available for the analysis. In univariate models, patients with excellent outcomes (mRS = 0-1) were younger, had higher TMT values and lower pre-event disability and stroke severity. In multivariate models higher TMT values resulted independently associated with reduced mortality (Odds Ratio 0.708, 95% Confidence Interval 0.538-0.930, p = 0.013). Age, diabetes, brain bleeding events and stroke severity were found to be predictors of mortality, too. Our retrospective analysis shows that in patients who underwent revascularization treatments for ischemic stroke TMT is as an independent predictor of survival easily obtainable from the baseline CT scan. Further investigation is required to confirm the role of sarcopenia assessment and TMT measurement in the prognostication toolkit of this disease.

Effectiveness of computed tomography perfusion imaging in stroke management.

Current guidelines do not support the use of computed tomography perfusion (CTP) in stroke, except when identifying the penumbra during an extended treatment window. Therefore, this study aimed to define the yield of CTP in diagnosing a stroke diagnosis beyond the imaging of the penumbra in the hyperacute phase (0-6 h) and an extended time window (6-24 h). All consecutive patients with acute onset of symptoms within a 24-h window underwent CTP imaging. The diagnostic value of CTP was calculated against the clinical and radiological diagnoses of stroke. A positive CTP result was determined by the presence of either a core or penumbra on the RAPID summary. Clinical diagnoses corresponded to discharge diagnoses of stroke. A radiological diagnosis was established if early ischemic changes [Alberta Stroke Program Early CT Score (ASPECTS) <10] were observed on the baseline CT scan, acute infarction was confirmed on follow-up imaging, or symptomatic occlusion was evident on baseline CTA. Between November 2018 and November 2019, 585 consecutive patients with an acute neurological deficit underwent multimodal CT imaging. A total of 500 patients (85%) were included, where 274 (55%) were within the hyperacute phase, 153 (31%) had a radiological diagnosis of stroke, and 122 (24%) had a clinical diagnosis of stroke. CTP yielded positive results only in patients with a confirmed stroke (positive predictive value and specificity of 100%). When CTP results were negative, 43% of the cases turned out to stroke mimics. Patients with stroke mimics were younger (66 ± 17 vs. 73 ± 13) and had lower scores on the National Institutes of Health Stroke Scale [median 0; interquartile range (IQR) 0-2 vs. median 4; IQR 2-6] compared to patients with CTP-negative strokes. In our study, CTP consistently indicated brain ischemia; therefore, in stroke management, CTP is most beneficial when it yields a positive result. A positive CTP result should prompt adequate stroke management actions without any delay. Conversely, a negative CTP result necessitates the consideration of both stroke and non-stroke diagnoses.

Body composition as a determinant of the therapeutic index with androgen signaling inhibition.

Androgen signaling is central to prostate cancer and men's health. Prior data indicates that increasing body fat is unfavorable in the localized setting yet associated with favorable outcomes in men with metastatic disease. Understanding the biological links between adiposity and prostate cancer may optimize the therapeutic index with ASI. We hypothesized that host adiposity and androgen synthesis are linked to the efficacy and toxicity of ASI for men with metastatic castration-resistant prostate cancer (mCRPC). A post-hoc analysis was done of NCT02703623 where men with mCRPC (n = 186) were treated for 8 weeks with abiraterone acetate, prednisone, and apalutamide (AAPA), and a satisfactory response was defined as a PSA decline >50%. Body composition was measured on baseline CT scans. Germline DNA WES was performed with a focus on variants in steroidogenic genes. Adipokine levels were measured in pre-treatment plasma. Germline polymorphisms in 3 genes involved in androgen synthesis (AKR1C3 rs12529, CYP17A1 rs6162, SRD5A2 rs523349) were associated with differences in body composition at baseline on ADT alone (prior to receipt of AAPA). Elevated subcutaneous adipose tissue index (SATi, p = 0.02), visceral adipose tissue index (VATi, p = 0.03), and BMI (p = 0.04) were associated with satisfactory response to AAPA. Leptin had positive correlation with VATi (r = 0.47) and SATi (r = 0.48). Inherited polymorphisms in androgen synthesis correlated with differences in body composition after exposure to ADT and warrant further investigation as candidate markers for body composition toxicity. Elevated subcutaneous and visceral adiposity were associated with improved response to ASI.

Unleashing deep features and radiomics to enhance best overall response prediction to immunotherapy in advanced non-small cell lung cancer.

e20611 Background: Predicting immunotherapy response in advanced non-small cell lung cancer (NSCLC) through non-invasive means could be a groundbreaking advancement. Deep features extracted from medical images by deep Convolutional Neural Networks (CNNs) along with radiomics have demonstrated promising results in different oncologic diseases. However, the method for extracting these deep features influences the predictive capabilities of the model. Methods: A retrospective analysis of a multicenter study was conducted in advanced NSCLC patients with high PD-L1 expression with the aim of developing an imaging features-based model for the prediction of best overall response (BOR) to a checkpoint inhibitor immunotherapy. All visible lung lesions, beyond RECIST 1.1, were manually segmented in baseline CT scans. Radiomics features were extracted at lesion-level, and various preprocessing methods and CNN architectures were tested for deep features extraction. Radiomics and deep features were harmonized using ComBat methodology. A distinctive feature signature was created for each patient by calculating the mean across all lesions. Using this data, machine learning models were trained to predict treatment response, specifically distinguishing between complete response (CR) and partial response (PR) from stable disease (SD) and progressive disease (PD). The study compared the performance of different deep features extraction methods and assessed shapley additive explanations (SHAP) values of radiomics for interpretability. Results: In the cohort of 116 patients (mean age: 67, 93% ever-smokers, 73% male), 1024 to 5475 quantitative imaging features per lesion were extracted. Among these patients, 67 patients were CR+PR, and 49 were SD+PD. The top-performing predictive model, trained with radiomics and deep features from a deep convolutional autoencoder, achieved an average AUC of 85%. SHAP values showed that impactful radiomics features were correlated with intensity changes in adjacent pixels describing the tumor heterogeneity, along with shape features potentially associated with malignancy. Ultimately, combining both features demonstrated statistical significance, leading to a performance improvement of ~2-5% compared to models exclusively using deep features. Conclusions: Diverse techniques for extracting deep features were evaluated, revealing autoencoder training as the most effective strategy for predicting immunotherapy response. Consequently, the findings suggest that employing deep learning to extract imaging features from baseline CT scans, combined with radiomics, represents a promising non-invasive approach for the predicting immunotherapy outcomes in advanced NSCLC patients.

CheckpointPx: A predictive radiology AI model of immune checkpoint inhibitor (ICI) benefit in non-small cell lung cancer (NSCLC).

8632 Background: Immune checkpoint inhibitors (ICIs) have dramatically transformed the field of non-small cell lung cancer (NSCLC) treatment. Despite the widespread use of PD-L1 as a biomarker in NSCLC, it has significant limitations as a reliable predictive biomarker for ICI response. Addressing this limitation, we have developed CheckpointPx, a non-invasive radiology AI tool to assist in ICI treatment selection for patients using only baseline CT scans. Here we demonstrate that CheckpointPx can predict improved outcomes specific to treatment with ICIs, presenting a new tool to address the shortage of reliable predictive ICI biomarkers in NSCLC and ultimately improve outcomes for patients undergoing immunotherapeutic interventions. Methods: CheckpointPx (v1.11) was trained on pre-treatment CT scans to predict response to ICIs from training data (D1: n=252 ICI recipients) from three institutions (A-C). In this study, we evaluated its performance in predicting response in two validation cohorts: a Treatment Dataset of patients receiving immunotherapy (D2: n=224, institutions B-D) and a Control Dataset of patients receiving platinum-based chemotherapy alone (D3: n=76, institution B). The response was defined as disease control per RECIST v1.1. Experienced radiologists and physicians delineated target lesions, and a deep learning model segmented pulmonary vessels. From these segmented regions, radiomic features were extracted using the Picture Health Px platform and were used to generate a radiomics benefit score and corresponding benefit groups using D1. The ability of benefit groups to stratify patients by progression-free survival (PFS) was compared across D2 and D3 to evaluate its utility in identifying patients who would benefit from ICI over chemotherapy. Results: D2 and D3 contained a mix of treatment lines (1st-4th) and predominantly late-stage tumors (Stage 3/4, &gt;85%). CheckpointPx included 16 features, such as measurements of intra-tumoral heterogeneity and vessel twistedness and branching patterns. Within the treatment dataset (D2), CheckpointPx was found to significantly stratify ICI patients by progression-free survival (PFS) with HR=0.68 (95% CI: 0.52 - 0.93, p=0.019). When applied to the control dataset, D3, benefit groups failed to stratify patients treated with chemotherapy by outcome (HR=0.91 [95% CI: 0.51-1.61], p=0.740), indicating that the signature was specific to ICI response. Conclusions: CheckpointPx demonstrated the ability to identify NSCLC patients who would benefit from ICI over chemo. The model’s association with PFS among ICI recipients, but not patients receiving chemotherapy alone, suggests that the signature is predictive of immunotherapy-related outcomes rather than generally prognostic. Additional independent, multi-site and prospective validation is warranted.

A rare cause of malignant otitis externa and skull base osteomyelitis.

An elderly male with type 2 diabetes presented with a 2-month history of otalgia and severe headaches. He was diagnosed with malignant otitis externa (MOE) and was commenced on empirical treatment with oral ciprofloxacin. Pseudomonas is the most common cause of MOE. A baseline CT scan was undertaken that demonstrated skull base osteomyelitis (SBO) due to findings of bone erosion at the mastoid tip and an infiltrating soft tissue mass eroding the clivus. Eight weeks later, he returned with worsening and bilateral symptoms of otitis externa, hearing loss, temporomandibular pain and dysfunction. Worsening and now bilateral malignant otitis externa were confirmed with an MRI scan that also demonstrated a small fluid collection in his left temporal region. The collection was aspirated and grew scedosporium apiospermum. He was diagnosed with fungal SBO and was commenced on treatment with the antifungal voriconazole, with significant improvement in symptoms and radiological findings. Fungal osteomyelitis is more likely in immunosuppressed patients, particularly those with type 2 diabetes. Fungal aetiology should be suspected in patients with progressive symptoms, despite treatment. A microbiology diagnosis of fungal SBO or MOE can be challenging to obtain and can lead to diagnostic delay. A sampling of the external auditory canal can aid in diagnosing MOE; however, scedosporium may also be isolated as a commensal organism. Aspirations from accessible fluid collections, infratemporal fossa needle sample and bone biopsy can provide material for diagnosis. Scedosporium is a rare cause of disease in humans, however, fungal infections are increasing in humans, due to an increase in susceptible populations. Scedosporium apiospermum is a rare cause of SBO and should be considered in patients not responding to standard treatment.

Effects of Achieving Rapid, Intensive, and Sustained Blood Pressure Reduction in Intracerebral Hemorrhage Expansion and Functional Outcome.

The time taken to achieve blood pressure (BP) control could be pivotal in the benefits of reducing BP in acute intracerebral hemorrhage (ICH). We aimed to assess the relationship between the rapid achievement and sustained maintenance of an intensive systolic BP (SBP) target with radiologic, clinical, and functional outcomes. Rapid, Intensive, and Sustained BP lowering in Acute ICH (RAINS) was a multicenter, prospective, observational cohort study of adult patients with ICH <6 hours and SBP ≥150 mm Hg at 4 Comprehensive Stroke Centers during a 4.5-year period. Patients underwent baseline and 24-hour CT scans and 24-hour noninvasive BP monitoring. BP was managed under a rapid (target achievement ≤60 minutes), intensive (target SBP <140 mm Hg), and sustained (target stability for 24 hours) BP protocol. SBP target achievement ≤60 minutes and 24-hour SBP variability were recorded. Outcomes included hematoma expansion (>6 mL or >33%) at 24 hours (primary outcome), early neurologic deterioration (END, 24-hour increase in NIH Stroke Scale score ≥4), and 90-day ordinal modified Rankin scale (mRS) score. Analyses were adjusted by age, sex, anticoagulation, onset-to-imaging time, ICH volume, and intraventricular extension. We included 312 patients (mean age 70.2 ± 13.3 years, 202 [64.7%] male). Hematoma expansion occurred in 70/274 (25.6%) patients, END in 58/291 (19.9%), and the median 90-day mRS score was 4 (interquartile range, 2-5). SBP target achievement ≤60 minutes (178/312 [57.1%]) associated with a lower risk of hematoma expansion (adjusted odds ratio [aOR] 0.43, 95% confidence interval [CI] 0.23-0.77), lower END rate (aOR 0.43, 95% CI 0.23-0.80), and lower 90-day mRS scores (aOR 0.48, 95% CI 0.32-0.74). The mean 24-hour SBP variability was 21.0 ± 7.6 mm Hg. Higher 24-hour SBP variability was not related to expansion (aOR 0.99, 95% CI 0.95-1.04) but associated with higher END rate (aOR 1.15, 95% CI 1.09-1.21) and 90-day mRS scores (aOR 1.06, 95% CI 1.04-1.10). Among patients with acute ICH, achieving an intensive SBP target within 60 minutes was associated with lower hematoma expansion risk. Rapid SBP reduction and stable sustention within 24 hours were related to improved clinical and functional outcomes. These findings warrant the design of randomized clinical trials examining the impact of effectively achieving rapid, intensive, and sustained BP control on hematoma expansion. This study provides Class III evidence that in adults with spontaneous ICH and initial SBP ≥150 mm Hg, lowering SBP to <140 mm Hg within the first hour and maintaining this for 24 hours is associated with decreased hematoma expansion.

Association of Acute Respiratory Disease Events with Quantitative Interstitial Abnormality Progression at CT in Individuals with a History of Smoking.

Background Acute respiratory disease (ARD) events are often thought to be airway-disease related, but some may be related to quantitative interstitial abnormalities (QIAs), which are subtle parenchymal abnormalities on CT scans associated with morbidity and mortality in individuals with a smoking history. Purpose To determine whether QIA progression at CT is associated with ARD and severe ARD events in individuals with a history of smoking. Materials and Methods This secondary analysis of a prospective study included individuals with a 10 pack-years or greater smoking history recruited from multiple centers between November 2007 and July 2017. QIA progression was assessed between baseline (visit 1) and 5-year follow-up (visit 2) chest CT scans. Episodes of ARD were defined as increased cough or dyspnea lasting 48 hours and requiring antibiotics or corticosteroids, whereas severe ARD episodes were those requiring an emergency room visit or hospitalization. Episodes were recorded via questionnaires completed every 3 to 6 months. Multivariable logistic regression and zero-inflated negative binomial regression models adjusted for comorbidities (eg, emphysema, small airway disease) were used to assess the association between QIA progression and episodes between visits 1 and 2 (intercurrent) and after visit 2 (subsequent). Results A total of 3972 participants (mean age at baseline, 60.7 years ± 8.6 [SD]; 2120 [53.4%] women) were included. Annual percentage QIA progression was associated with increased odds of one or more intercurrent (odds ratio [OR] = 1.29 [95% CI: 1.06, 1.56]; P = .01) and subsequent (OR = 1.26 [95% CI: 1.05, 1.52]; P = .02) severe ARD events. Participants in the highest quartile of QIA progression (≥1.2%) had more frequent intercurrent ARD (incidence rate ratio [IRR] = 1.46 [95% CI: 1.14, 1.86]; P = .003) and severe ARD (IRR = 1.79 [95% CI: 1.18, 2.73]; P = .006) events than those in the lowest quartile (≤-1.7%). Conclusion QIA progression was independently associated with higher odds of severe ARD events during and after radiographic progression, with higher frequency of intercurrent severe events in those with faster progression. Clinical trial registration no. NCT00608764 © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Little in this issue.

Quantitative Analysis of Lung Shape in Idiopathic Pulmonary Fibrosis: Insights Into Disease- and Age-Associated Patterns

Rationale and objectivesFibrotic scarring in idiopathic pulmonary fibrosis (IPF) typically develops first in the posterior-basal lung tissue before advancing to involve more of the lung. The complexity of lung shape in the costo-diaphragmatic region has been proposed as a potential factor in this regional development. Intrinsic and disease-related shape could therefore be important for understanding IPF risk and its staging. We hypothesised that lung and lobe shape in IPF would have important differences from controls. Materials and MethodsA principal component (PC) analysis was used to derive a statistical shape model (SSM) of the lung for a control cohort aged > 50 years (N=39), using segmented lung and fissure surface data from CT imaging. Individual patient shape models derived for baseline (N=18) and follow-up (N=16) CT scans in patients with IPF were projected to the SSM to describe shape as the sum of the SSM average and weighted PC modes. Associations between the first four PC shape modes, lung function, percentage of fibrosis (fibrosis%) and pulmonary vessel-related structures (PVRS%), and other tissue metrics were assessed and compared between the two cohorts. ResultsShape was different between IPF and controls (p<0.05 for all shape modes), with IPF shape forming a distinct shape cluster. Shape had a negative relationship with age in controls (p = 0.013), but a positive relationship with age in IPF (p = 0.026). Some features of shape changed on follow-up. Shape in IPF was associated with fibrosis% (p<0.05) and PVRS% (p<0.05). ConclusionQuantitative comparison of lung and lobe shape in IPF with controls of a similar age reveals shape differences that are strongly associated with age and percent fibrosis. The clustering of IPF cohort shape suggests that it could be an important feature to describe disease.

Abstract 912: CheckpointPx, an interpretable radiology AI tool, predicts checkpoint blockade benefit independent of PDL1 status in non-small cell lung cancers (NSCLC): A multi-institutional validation study

Abstract Background: Immune Checkpoint Inhibitors (ICIs) are a cornerstone in the treatment of cancers such as Non-Small Cell Lung Cancer (NSCLC). Despite the standard use of biomarkers such as PD-L1 expression for selecting ICI regimens, recent studies suggest missed opportunities for benefit in PD-L1 low/negative patients. Therefore, there is an urgent need for new biomarker strategies capable of guiding the use of ICIs in NSCLC. We introduce and validate an AI-powered radiology tool for ICI patient selection that provides interpretable predictions by quantifying aspects of tumor biology on baseline CT scans directly associated with ICI efficacy, such as tumor heterogeneity and twistedness of the tumor-associated vasculature. Methods: We analyzed CT scans of 439 patients treated with ICI at 4 institutions (D1-D4) using the Picture Health Px platform. Experienced radiologists and physicians across institutions delineated target lesions on baseline CT scans. A deep learning model was used to segment pulmonary vessels. CheckpointPx v1.11, an interpretable neural network incorporating quantitative features extracted from within the tumor and its vasculature, was trained to predict ICI response (defined as disease control per RECIST best overall response) on n=247 patients (D1-D3). It was assessed with respect to response and progression free-survival (PFS) on 192 patients (D2-D4), with D4 (n=105) being an external validation set. CheckpointPx was also evaluated within PDL1 subsets among testing set patients where available (n=138). Results: The cohort was predominantly late stage (Stage 3/4) (&amp;gt;85%) and was mixed with respect to ICI line of treatment (1st-4th). 122 of 247 (49%) and 119 of 192 (62%) were responders in training and test cohorts respectively. CheckpointPx included 16 features, such as entropy-based heterogeneity and quantitative vessel tortuosity (twistedness). The model predicted response to ICI with an AUC=0.65 on the test set. The CheckpointPx High Risk group significantly stratified patients by PFS (HR=1.67 [1.22-2.28], p=0.001). This separation remained significant within the subset of PDL1-negative (HR=2.71 [1.35-5.44], p=0.005) and PDL1-positive patients (HR=2.05 [1.22-3.46], p=0.007). Conclusions: From baseline radiology, CheckpointPx was shown to strongly predict ICI outcomes across multiple institutions, NSCLC stages, and lines of ICI. Furthermore, the tool stratified patients by ICI benefit within both PDL1-positive and negative subsets - suggesting the potential of CheckpointPx to address critical gaps in the NSCLC biomarker landscape. CheckpointPx is driven by interpretable imaging biomarkers, and thus its predictions are tied to specific and quantifiable phenotypic attributes of the tumor and its microenvironment. Citation Format: Amogh Hiremath, Seyoung Lee, Jeeyeon Lee, Peter Haseok Kim, Kai Zhang, Salie Lee, Monica Yadav, Maria Jose Aguilera Chuchuca, Taegyu Um, Myungwoo Nam, Liam Il-Young Chung, Hye Sung Kim, Jisang Yu, Trie Arni Djunadi, Leeseul Kim, Youjin Oh, Sungmi Yoon, Zunairah Shah, Yuchan Kim, Ilene Hong, Grace Kang, Jessica Jang, Amy Cho, Soowon Lee, Cecilia Nam, Timothy Hong, Yury S. Velichko, Amit Gupta, Vamsidhar Velcheti, Anant Madabhushi, Nathaniel Braman, Young Kwang Chae. CheckpointPx, an interpretable radiology AI tool, predicts checkpoint blockade benefit independent of PDL1 status in non-small cell lung cancers (NSCLC): A multi-institutional validation study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 912.

Abstract 2594: AI-powered radiomics model predicts immune checkpoint inhibitor-related pneumonitis (CIP) in advanced NSCLC patients

Abstract Background: With the escalating integration of immunotherapy in the management of advanced non-small cell lung cancer (NSCLC), the emergence of adverse events, particularly immune checkpoint inhibitor-related pneumonitis (CIP), poses important challenges. CIP is not uncommon and can be life-threatening. It often necessitates the discontinuation of immunotherapy, even in patients with an otherwise favorable response. Prevention, early detection and early management of CIP can enhance patient outcomes, yet no such predictive models have been established. This study investigates the use of Artificial Intelligence (AI) algorithms in analyzing radiomic features for the prediction of CIP in NSCLC patients receiving immunotherapy. Methods: A cohort of 105 stage III-IV NSCLC patients receiving immunotherapy was examined. Half of the patients were randomly split to a training set, while the remaining half were reserved for algorithm testing. The manual segmentation was performed by three physicians annotating in consensus using LIFEx software v7.3.0 (IMIV/CEA, Orsay, France). The Picture Health Px platform was utilized to perform an AI-powered deep phenotyping of the tumor and its surrounding habitat. A number of interpretable feature measures were extracted from baseline CT scans, which were in turn used to train a deep learning classifier for the detection of pneumonitis. Weighting techniques were applied to compensate for the imbalance of pneumonitis cases. Results: Among the 105 patients, 63 (60.0%) received immunotherapy-only regimen and 42 (40.0%) received combination immunochemotherapy. 18 (17.1%) patients had pneumonitis events. Within this subset, 10 (55.6%) had CIP. Among the CIP group, six patients (60.0%) had grade 1 pneumonitis, three patients (30.0%) had grade 2 pneumonitis, and one patient (10.0%) had grade 3 pneumonitis, and none had grade 4 or grade 5 pneumonitis. Within the training set (n=51), the cross-validated area under the ROC curve (AUC) was 0.71 (95% CI: 0.68-0.74). When applied to the test set, the model predicted pneumonitis with AUC=0.63. Across the two datasets, the model correctly identified 4/6 (66.7%) grade 1 pneumonitis events and 2/3 (66.7%) grade 2 pneumonitis events, but misclassified the only available grade 3 event. Conclusion: The utilization of CT-based radiomic features demonstrates promise in predicting CIP in NSCLC patients undergoing immunotherapy. This approach holds potential for enhancing the identification and management of CIP, among NSCLC patients treated with immunotherapy. Citation Format: Seyoung Lee, Amogh Hiremath, Jeeyeon Lee, Peter Haseok Kim, Kai Zhang, Salie Lee, Monica Yadav, Maria J. Chuchuca, Taegyu Um, Myungwoo Nam, Liam Il-Young Chung, Hye Sung Kim, Jisang Yu, Trie Arni Djunadi, Leeseul Kim, Youjin Oh, Sungmi Yoon, Zunairah Shah, Yuchan Kim, Ilene Hong, Grace Kang, Jessica Jang, Amy Cho, Soowon Lee, Cecilia Nam, Timothy Hong, Yuri S. Velichko, Vamsidhar Velcheti, Anant Madabhushi, Nathaniel Braman, Young Kwang Chae. AI-powered radiomics model predicts immune checkpoint inhibitor-related pneumonitis (CIP) in advanced NSCLC patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2594.

Optimal intraventricular hemorrhage volume cutoff for predicting poor outcome in patients with intracerebral hemorrhage

Background and objectivesThe prognosis of patients with spontaneous intracerebral hemorrhage (ICH) is often influenced by hematoma volume, a well-established predictor of poor outcome. However, the optimal intraventricular hemorrhage (IVH) volume cutoff for predicting poor outcome remains unknown. MethodsWe analyzed 313 patients with spontaneous ICH not undergoing evacuation, including 7 cases with external ventricular drainage (EVD). These patients underwent a baseline CT scan, followed by a 24-hour CT scan for measurement of both hematoma and IVH volume. We defined hematoma growth as hematoma growth > 33 % or 6 mL at follow-up CT, and poor outcome as modified Rankin Scale score≥3 at three months. Cutoffs with optimal sensitivity and specificity for predicting poor outcome were identified using receiver operating curves. ResultsThe receiver operating characteristic analysis identified 6 mL as the optimal cutoff for predicting poor outcome. IVH volume> 6 mL was observed in 53 (16.9 %) of 313 patients. Patients with IVH volume>6 mL were more likely to be older and had higher NIHSS score and lower GCS score than those without. IVH volume>6 mL (adjusted OR 2.43, 95 % CI 1.13-5.30; P = 0.026) was found to be an independent predictor of poor clinical outcome at three months in multivariable regression analysis. ConclusionsOptimal IVH volume cutoff represents a powerful tool for improving the prediction of poor outcome in patients with ICH, particularly in the absence of clot evacuation or common use of EVD. Small amounts of intraventricular blood are not independently associated with poor outcome in patients with intracerebral hemorrhage. The utilization of optimal IVH volume cutoffs may improve the clinical trial design by targeting ICH patients that will obtain maximal benefit from therapies.

Significant Spontaneous Pneumomediastinum and Extensive Subcutaneous Emphysema in a COVID-19 Patient

A 64-year-old man, who had no pre-existing health conditions, was admitted to the intensive care unit due to progressive shortness of breath resulting from COVID-19. Initially, the patient responded to non-invasive mechanical ventilation, which improved his breathing function. However, after six days, his respiratory function worsened significantly, requiring invasive ventilation. Out of nowhere, the person experienced spontaneous pneumomediastinum and extensive subcutaneous emphysema. The next day, a spontaneous pneumothorax occurred and was successfully drained later. It became evident that there was extensive subcutaneous emphysema also. The CT scan of the chest confirmed the presence of spontaneous pneumomediastinum, but it did not have any effect on the patient’s hemodynamics. The physicians performed a bronchoscopy and imaging with a contrast medium, which ruled out any lesions to the trachea or esophagus. No other issues related to the problem were identified during the examination. Unfortunately, microscopic bullae of interstitial emphysema, or micro air leaks, are visible when they are an adequate size on computed tomography. Recent literature and studies not available during the case report have shown that if the Macklin effect is detected on the baseline CT scan, it could predict the occurrence of pneumothorax or pneumomediastinum.