Single Chest Research Articles

233. UTILITY OF SINGLE TRANSHIATAL CHEST DRAINAGE AFTER MINIMALLY INVASIVE OESOPHAGECTOMY: A PROPENSITY-SCORE MATCHED STUDY OF FEASIBILITY

Abstract Background The routine placement of intercostal chest drains remains common practice following minimally invasive oesophagectomy (MIO). Chest drains may contribute to increased post-operative pain and morbidity. The aim of this study was to compare the outcomes of transhiatal chest drain with intercostal chest drain. We analysed use of analgesia, length of hospital stay, and postoperative pulmonary morbidity following MIO. Methods This was a retrospective case-control study of patients undergoing elective MIO at a quaternary centre in Western Australia between September 2016 and September 2023. To reduce confounding, propensity score matching was performed. All patients were identified from a prospectively maintained database. In patients undergoing MIO with exclusive transhiatal drainage, a 19 French Blake’s drain was advanced through the hiatus into the right pleural cavity and positioned adjacent to the anastomosis. Results Thirty patients were included in the study, fifteen of whom had exclusively transhiatal drain placement. Mean total daily oral morphine dose equivalent was significantly lower in the transhiatal drain group (58mg+/-35.8) compared with intercostal drain group (100.8mg+/-64.0) on day 1 (p=0.04). There is no statistically significant difference between the total oral morphine use on days 3 (p=0.44) and 6 (p=0.97). No significant difference was observed in reintervention rate (p=1.00), length of stay (p=0.78) or postoperative pulmonary complications (p=1.00). Conclusion The results of this study suggest that transhiatal drain placement following MIO is associated with a reduction in post-operative pain and no increase in pulmonary complications or length of stay. Larger, prospective studies are needed to further evaluate the safety of single transhiatal chest drain placement following MIO.

Predictive model for intraoperative decision-making in video-assisted thoracoscopic lobectomy: optimizing chest drain placement for high-risk patients.

For years, it has been the common and widely accepted practice in thoracic surgery to place apical and basal drains after a lobectomy to completely drain the pleural cavity. With the development of thoracoscopic technology, it became apparent that the use of a single chest tube provided the same clinical results. However, sometimes tension pneumothorax occurs with the need for additional pleural drainage. The aim of this study was to develop a prognostic model to identify high risk patients intraoperatively and to insert additional pleural drainage to prevent the development of pneumothorax after video-assisted thoracoscopic surgery (VATS) lobectomy. This was a retrospective multicenter study of patients (registry data was analyzed) who underwent VATS lobectomy via a standardized multiport approach between 2014 and 2022. In all cases, a single drain was used postoperatively. We used a machine learning algorithm and data synthesis to expand patient selection according to Riley's method. A total of 418 cases were analyzed in this study. After determining the prognostically significant factors, we performed a binary logistic regression analysis using reverse step-by-step inclusion of variables according to the Akaike information criterion. After validation of the model by bootstrap (400 iterations) and with the original dataset, a nomogram with a specific point distribution for each risk factor was created. The rate of tension pneumothorax was 4.53% (n=66). The most significant variables associated with the need for additional drainage were adhesions, intraoperative lung suturing, fused interlobar fissure, enlarged intrapulmonary lymph nodes, chronic obstructive pulmonary disease (P<0.001). The C-index of the model was 0.957, the mean absolute calibration error was 0.6%, and the slope of the calibration curve was 0.959. A score of 26 points indicated a 95% risk of postoperative tension pneumothorax. The nomogram achieved good predictive performance for tension pneumothorax after minimally invasive lobectomy. High-risk patients could be identified, and additional drainage may be placed intraoperatively to reduce the risk of lung collapse in the postoperative period.

Outcomes of a Single Transverse Chest Roll for Prone Positioning Technique During Percutaneous Nephrolithotomy

ObjectiveTo compare anesthetic parameters using a novel prone single transverse chest roll technique (STR) to the standard thoraco-pelvic dual transverse roll technique (DTR). MethodsA retrospective review of 441 patients who underwent PCNL between 2018 to 2022 was performed. A total of 4 surgeons were included—surgeon 1 utilized the STR technique while surgeons 2, 3, and 4 used the DTR technique. Anesthetic parameters including end tidal CO2 (ETCO2), mean arterial pressure (MAP), peak airway pressure (Ppeak), plateau airway pressure (Pplat), positive end expiratory pressure (PEEP), oxygen saturation (SpO2), and tidal volume (TV) were compared between both groups at 0 (supine), 15, 30, and 60 minutes post-intubation intervals. Mixed effects regression models with interaction and pairwise comparisons were made between both groups (p<0.05). ResultsA total of 581 PCNLs were performed with 199 using STR and 382 using DTR. Surgery duration, ASA class, and age were similar amongst the STR and DTR groups. Estimated blood loss (59cc vs 83cc, p=0.007) and length of stay (77 hrs vs 163 hrs, p=<0.001) was significantly lower in the STR group. There was a significantly lower Ppeak, Pplat and TV in the STR compared to DTR group at 0, 15, 30, and 60 minutes (p<0.001). ConclusionUsage of a single transverse chest roll during prone PCNL appears to be a safe positioning method. STR patients had lower Ppeak and Pplat at all timepoints, which has been shown to be predictive of lower blood loss.

Ultrafiltration of Angelicae Sinensis Radix and Astragali Radix inhibits ferroptosis and improves pulmonary fibrosis in rats by regulating Nrf2/xCT/GPX4 signaling pathway

This study aims to investigate the mechanism of ferroptosis mediated by the nuclear factor-E2-related factor 2(Nrf2)/solute carrier family 7 member 11(SLC7A11, also known as xCT)/glutathione peroxidase 4(GPX4) signaling pathway in radiationinduced pulmonary fibrosis and the intervention effect of Angelicae Sinensis Radix(ASR) and Astragali Radix(AR) ultrafiltration extract. Fifty Wistar rats were randomly divided into five groups, with 10 rats in each group. Except for the blank group without radiation, the rats in each group were anesthetized and subjected to a single local chest irradiation of 40 Gy X-rays once to establish a rat model of radiation-induced pulmonary fibrosis. After radiation, the rats in the intervention groups were orally administered with ASR-AR ultrafiltration extract at doses of 0. 12, 0. 24, and 0. 48 g·kg~(-1), respectively, once a day for 30 days. After 30 days of continuous administration, the levels of oxidative stress indicators superoxide dismutase(SOD) activity, reduced glutathione(GSH),malondialdehyde(MDA), and ferrous ion(Fe~(2+)) in lung tissues of each group were detected by colorimetry. Immunofluorescence was used to detect reactive oxygen species(ROS) fluorescence expression in lung tissues. Hematoxylin-eosin(HE) and Masson staining were performed to observe pathological changes in lung tissues. Immunohistochemistry and Western blot were used to detect the expression levels of Nrf2/xCT/GPX4 signaling pathway and fibrotic proteins in lung tissues. The results showed that compared with the results in the blank group, the levels of Fe~(2+) and MDA in the model group increased, while SOD activity and GSH levels decreased,and ROS levels increased. HE and Masson staining results showed that the structure of lung tissue was seriously damaged, the pulmonary interstitium was significantly proliferated, the alveoli collapsed and consolidated severely, and there were more inflammatory cell aggregates and collagen fiber deposits. Transmission electron microscopy showed that the degree of lung tissue damage in the model group was relatively high, with increased, smaller, and disorganized damaged mitochondria, irregular morphology, shallow matrix,most mitochondria ruptured and shortened, mildly expanded, some mitochondria with increased electron density of the matrix, partial mitochondrial outer membrane rupture, and characteristic changes of ferroptosis-specific mitochondria. Immunohistochemistry showed that the expression of transferrin receptor protein 1(TFR1) in lung tissues was significantly increased, while the expression of GPX4,ferritin heavy chain 1(FTH1), Nrf2, and xCT was significantly decreased. Western blot showed that the expression of α-smooth muscle actin(α-SMA) and collagen Ⅰ protein increased. Compared with the model group, the intervention group with ASR-AR ultrafiltration extract significantly improved lipid peroxidation and antioxidant-related indicators, decreased Fe~(2+) levels, alleviated fibrosis, and decreased the expression of TFR1, α-SMA, and collagen Ⅰ proteins in lung tissues, while increased the expression of GPX4, FTH1, Nrf2, and xCT proteins. In summary, ASR-AR ultrafiltration extract has an ameliorative effect on radiation-induced pulmonary fibrosis, and its mechanism may involve the inhibition of ferroptosis by regulating the Nrf2/xCT/GPX4 signaling pathway.

Measuring gait parameters from a single chest-worn accelerometer in healthy individuals: a validation study

Digital health technologies (DHTs) are increasingly being adopted in clinical trials, as they enable objective evaluations of health parameters in free-living environments. Although lumbar accelerometers notably provide reliable gait parameters, embedding accelerometers in chest devices, already used for vital signs monitoring, could capture a more comprehensive picture of participants’ wellbeing, while reducing the burden of multiple devices. Here we assess the validity of gait parameters measured from a chest accelerometer. Twenty healthy adults (13 females, mean ± sd age: 33.9 ± 9.1 years) instrumented with lumbar and chest accelerometers underwent in-lab and outside-lab walking tasks, while monitored with reference devices (an instrumented mat, and a 6-accelerometers set). Gait parameters were extracted from chest and lumbar accelerometers using our open-source Scikit Digital Health gait (SKDH-gait) algorithm, and compared against reference values via Bland–Altman plots, Pearson’s correlation, and intraclass correlation coefficient. Mixed effects regression models were performed to investigate the effect of device, task, and their interaction. Gait parameters derived from chest and lumbar accelerometers showed no significant difference and excellent agreement across all tasks, as well as good-to-excellent agreement and strong correlation against reference values, thus supporting the deployment of a single multimodal chest device in clinical trials, to simultaneously measure gait and vital signs.Trial Registration: The study was reviewed and approved by the Advarra IRB (protocol number: Pro00043100).

Multi-scale Lesion Feature Fusion and Location-Aware for Chest Multi-disease Detection.

Accurately identifying and locating lesions in chest X-rays has the potential to significantly enhance diagnostic efficiency, quality, and interpretability. However, current methods primarily focus on detecting of specific diseases in chest X-rays, disregarding the presence of multiple diseases in a single chest X-ray scan. Moreover, the diversity in lesion locations and attributes introduces complexity in accurately discerning specific traits for each lesion, leading to diminished accuracy when detecting multiple diseases. To address these issues, we propose a novel detection framework that enhances multi-scale lesion feature extraction and fusion, improving lesion position perception and subsequently boosting chest multi-disease detection performance. Initially, we construct a multi-scale lesion feature extraction network to tackle the uniqueness of various lesion features and locations, strengthening the global semantic correlation between lesion features and their positions. Following this, we introduce an instance-aware semantic enhancement network that dynamically amalgamates instance-specific features with high-level semantic representations across various scales. This adaptive integration effectively mitigates the loss of detailed information within lesion regions. Additionally, we perform lesion region feature mapping using candidate boxes to preserve crucial positional information, enhancing the accuracy of chest disease detection across multiple scales. Experimental results on the VinDr-CXR dataset reveal a 6% increment in mean average precision (mAP) and an 8.4% improvement in mean recall (mR) when compared to state-of-the-art baselines. This demonstrates the effectiveness of the model in accurately detecting multiple chest diseases by capturing specific features and location information.

Accuracy of Automatic Chest Compression Detection of Different Manufacturers

Research question The detection of single chest compressions (CCs) in defibrillator records is crucial to evaluate CPR quality parameters like longest pause duration.1 Currently, defibrillators detect CCs automatically and provide performance feedback via their proprietary software. While some manufacturers (e.g. Stryker) report the accuracy of their CC detection algorithm and allow to manually revise the automatically detected CCs to improve accuracy, others (e.g. ZOLL) do not offer this option.2 Recent works further suggest that using automatically detected CCs without revision or other open source methods is sufficient to compute quality markers.3,4 We aim to compare the accuracy of the automatic CC detection of two defibrillators. Methodology 131 defibrillator recordings from ZOLL’s X-Series devices with an applied feedback sensor and 70 recordings of Stryker’s LIFEPAK 15 devices were exported. ZOLL detects CC based on accelerometry data from its CC feedback sensor, while Stryker uses the thoracic impedance signal. Each set of recordings was annotated by a single annotator by adding missing CCs and deleting excess CCs, forming a ground truth. The results are reported as median, (10th percentile, 90th percentile) and were tested on statistical significance with a Mann-Whitney U test. Results Per case, the device by ZOLL detects in median 99.6, (97.8,99.9) % of all CCs correctly. 0.4, (0.1,2.3) % are deleted and 0.4, (0.1,2.3) % are added during the annotation process. For Stryker’s LIFEPAK 15 the respective numbers are: correctly detected: 96.7, (81.5,99.2) %, deleted: 1.8, (0.2,10.9) %, added: 3.3, (.8,18.5) %. The difference between the correctly identified CCs is significant (p&lt;0.0001). The distribution of missing and excess CCs for all cases is shown in Figure 1. Interpretation It appears that ZOLL’s CC detection via an accelerometry based feedback sensor is more accurate than Stryker’s method using thoracic impedance. However, Stryker’s accuracy exceeds 95% as well, providing a reasonably reliable basis for CPR quality marker calculations.

Spirometry test values can be estimated from a single chest radiograph.

Physical measurements of expiratory flow volume and speed can be obtained using spirometry. These measurements have been used for the diagnosis and risk assessment of chronic obstructive pulmonary disease and play a crucial role in delivering early care. However, spirometry is not performed frequently in routine clinical practice, thereby hindering the early detection of pulmonary function impairment. Chest radiographs (CXRs), though acquired frequently, are not used to measure pulmonary functional information. This study aimed to evaluate whether spirometry parameters can be estimated accurately from single frontal CXR without image findings using deep learning. Forced vital capacity (FVC), forced expiratory volume in 1 s (FEV1), and FEV1/FVC as spirometry measurements as well as the corresponding chest radiographs of 11,837 participants were used in this study. The data were randomly allocated to the training, validation, and evaluation datasets at an 8:1:1 ratio. A deep learning network was pretrained using ImageNet. The input and output information were CXRs and spirometry test values, respectively. The training and evaluation of the deep learning network were performed separately for each parameter. The mean absolute error rate (MAPE) and Pearson's correlation coefficient (r) were used as the evaluation indices. The MAPEs between the spirometry measurements and AI estimates for FVC, FEV1 and FEV1/FVC were 7.59% (r = 0.910), 9.06% (r = 0.879) and 5.21% (r = 0.522), respectively. A strong positive correlation was observed between the measured and predicted indices of FVC and FEV1. The average accuracy of >90% was obtained in each estimation of spirometry indices. Bland-Altman analysis revealed good agreement between the estimated and measured values for FVC and FEV1. Frontal CXRs contain information related to pulmonary function, and AI estimation performed using frontal CXRs without image findings could accurately estimate spirometry values. The network proposed for estimating pulmonary function in this study could serve as a recommendation for performing spirometry or as an alternative method, suggesting its utility.

Ameliorated chest drain wound closure in patients undergoing uniportal thoracoscopic pulmonary resection.

Although uniportal video-assisted thoracoscopic surgery (VATS) has been performed for a wide array of thoracic diseases, unsightliness and poor wound healing often occur, particularly when a chest drain is placed postoperatively. Different chest drain wound closure (CWC) methods have been introduced with the benefits of cosmesis and patient satisfaction. We aimed to describe our improved CWC technique in this setting and assess its efficacy. A total of consecutive 334 patients undergoing uniportal VATS pulmonary resection with single chest drain placement were investigated from 2016 to 2021. The techniques for CWC were classified into the conventional method (35 patients, group A), continuous suture with removal-free stitches (122 patients, group B), and continuous suture with removal-free barbed suture plus topical skin adhesives (177 patients, group C). Perioperative data and complications related to CWC were analyzed. Group C had a significantly shorter operative time, postoperative hospital stay, and chest tube days than groups A and B (all p < 0.01). In terms of chest tube-related complications, there were no statistically significant differences in post-removal pneumothorax, subcutaneous emphysema, incisional effusion leakage, wound dehiscence, or infection. Overall, significant differences in scar scale scores were observed between the groups, where the ameliorated group C was superior to the conventional group A (p < 0.01). The improved CWC technique using continuous sutures with removal-free barbed sutures and topical skin adhesives is simple, safe, and effective. This may be a favorable CWC strategy when performing uniportal VATS, with enhanced patient satisfaction.

MA19.03 Sex Differences in the Prediction of Future Lung Cancer Risk Based on a Single Low-Dose Chest Computed Tomography Scan

MA19.03 Sex Differences in the Prediction of Future Lung Cancer Risk Based on a Single Low-Dose Chest Computed Tomography Scan

Accidental Self-inflicted Nail-gun Injury to the Heart

Abstract Penetrating cardiac injuries are heart crises requiring immediate surgical intervention, and nail-gun injuries have also been documented. This case report covers a 19-year-old male who self-inflicted a single chest wound using a nail gun. Examination of the heart revealed a nail that touched the myocardium around the right ventricular outflow tract.

Estimation of Lifetime Attributable Risk (LAR) of Cancer Associated with Chest Computed Tomography Procedures in Children

Purpose: The danger of radiation at low doses continues linearly, and without a threshold, investigations concluded that although the risk of cancer from Computed Tomography (CT) scans is low, it is not zero.&#x0D; This study aims to determine the patient's radiation dose and estimate the Lifetime Attributable Risk (LAR) of cancer incidence for a single chest CT scan in children.&#x0D; Materials and Methods: We divided 1,105 children into four age groups: 0 years, 5 years, 10 years, and 15 years. Dosimetric data of chest CT scan were plugged in VirtualDoseCT software, and organ dose and effective dose were calculated. The cancer risk based on organ dose is estimated according to the BEIR VII report.&#x0D; Results: The highest dose in boys was related to lung (5.13 - 6.8 mSv) and heart (5.27-5.97 mSv), and in girls, lung (4.98 - 5.91 mSv), breast (4.24 - 5.21 mSv), and heart (4.9 - 5.71 mSv) had the highest dose. The highest LAR (per 100,000) was obtained for the breast in the age group of 0 years (61.01), followed by the breast for the age group of 5 years (46.16) and lung in the age group of 0 years (43.32) in girls.&#x0D; Conclusion: This study shows a better concept of radiation dose in the chest CT scan in children and how much effective dose and organ dose values increase the cancer risk.

A Wireless Wearable Patch for Remote Monitoring of Respiratory and Cardiac Rhythm in Children.

Remote patient monitoring (RPM) is an innovative strategy to promote health and improve patient management and care. Recent advances in healthcare technologies have seen the emergence of wearable sensors allowing longitudinal physiological measurements in any environment. This paper introduces a wireless wearable patch 'Leo' for continuous remote monitoring of physiological data at home and healthcare settings. This includes single lead ECG, chest impedance, heart rate (HR), respiration rate (RR) and body posture. To test Leo's ability to capture longitudinal physiological data at home, 15 children experiencing acute severe asthma exacerbations were recruited during their emergency department (ED) visits. Participants wore the Leo device for 7 (+/-2) days post-hospital discharge. Nocturnal RR and HR and variability were higher during the first half of the night on Day1 compared to Day7 (p<0.005). Participants also completed a usability questionnaire and reported the patch wear to be comfortable (average score of 3.3 out of 5) and easy to wear during the night (average score of 3.5 out of 5) with 5/15 (33%) reported very slight barely perceptible skin irritation/redness and 2 (13%) reported well defined skin irritation and redness.Clinical Relevance- These results highlight the potential use of the Leo device in clinical practice for continuous un-obstructive monitoring of diseased populations, such as asthma.

Single chest drain is not inferior to double chest drain after robotic esophagectomy: a propensity score-matched analysis.

Chest drain management has a significant influence on postoperative recovery after robot-assisted minimally invasive esophagectomy (RAMIE). The use of chest drains increases postoperative pain by irritating intercostal nerves and hinders patients from early postoperative mobilization and recovery. To our knowledge, no study has investigated the use of two vs. one intercostal chest drains after RAMIE. This retrospective cohort study evaluated patients undergoing elective RAMIE with gastric conduit pull-up and intrathoracic anastomosis. Patients were divided into two groups according to placement of one (11/2020-08/2022) or two (08/2018-11/2020) chest drains. Propensity score matching was performed in a 1:1 ratio, and the incidences of overall and pulmonary complications, drainage-associated re-interventions, radiological diagnostics, analgesic use, and length of hospital stay were compared between single drain and double drain groups. During the study period, 194 patients underwent RAMIE. Twenty-two patients were included after propensity score matching in the single and double chest drain group, respectively. Time until removal of the last chest drain [postoperative day (POD) 6.7 ± 4.4 vs. POD 9.4 ± 2.7, p = 0.004] and intensive care unit stay (4.2 ± 5.1 days vs. 5.3 ± 3.5 days, p = 0.01) were significantly shorter in the single drain group. Overall and pulmonary complications, drainage-associated events, re-interventions, number of diagnostic imaging, analgesic use, and length of hospital stay were comparable between both groups. This study is the first to demonstrate the safety of single intercostal chest drain use and, at least, non-inferiority to double chest drains in terms of perioperative complications after RAMIE.

Chest Radiography of Tuberculosis: Determination of Activity Using Deep Learning Algorithm.

Inactive or old, healed tuberculosis (TB) on chest radiograph (CR) is often found in high TB incidence countries, and to avoid unnecessary evaluation and medication, differentiation from active TB is important. This study develops a deep learning (DL) model to estimate activity in a single chest radiographic analysis. A total of 3,824 active TB CRs from 511 individuals and 2,277 inactive TB CRs from 558 individuals were retrospectively collected. A pretrained convolutional neural network was fine-tuned to classify active and inactive TB. The model was pretrained with 8,964 pneumonia and 8,525 normal cases from the National Institute of Health (NIH) dataset. During the pretraining phase, the DL model learns the following tasks: pneumonia vs. normal, pneumonia vs. active TB, and active TB vs. normal. The performance of the DL model was validated using three external datasets. Receiver operating characteristic analyses were performed to evaluate the diagnostic performance to determine active TB by DL model and radiologists. Sensitivities and specificities for determining active TB were evaluated for both the DL model and radiologists. The performance of the DL model showed area under the curve (AUC) values of 0.980 in internal validation, and 0.815 and 0.887 in external validation. The AUC values for the DL model, thoracic radiologist, and general radiologist, evaluated using one of the external validation datasets, were 0.815, 0.871, and 0.811, respectively. This DL-based algorithm showed potential as an effective diagnostic tool to identify TB activity, and could be useful for the follow-up of patients with inactive TB in high TB burden countries.

Performance of radiomics models derived from different CT reconstruction parameters for lung cancer risk prediction

BackgroundThis study analysed the performance of radiomics features extracted from computed tomography (CT) images with different reconstruction parameters in differentiating malignant and benign pulmonary nodules.MethodsWe evaluated routine chest CT images acquired from 148 participants with pulmonary nodules, which were pathologically diagnosed during surgery in West China Hospital, including a 5 mm unenhanced lung window, a 5 mm unenhanced mediastinal window, a 5 mm contrast-enhanced mediastinal window and a 1 mm unenhanced lung window. The pulmonary nodules were segmented, and 1409 radiomics features were extracted for each window. Then, we created 15 cohorts consisting of single windows or multiple windows. Univariate correlation analysis and principal component analysis were performed to select the features, and logistic regression analysis was performed to establish models for each cohort. The area under the curve (AUC) was applied to compare model performance.ResultsThere were 75 benign and 73 malignant pulmonary nodules, with mean diameters of 18.63 and 19.86 mm, respectively. For the single-window setting, the AUCs of the radiomics model from the 5 mm unenhanced lung window, 5 mm unenhanced mediastinal window, 5 mm contrast-enhanced mediastinal window and 1 mm unenhanced lung window were 0.771, 0.808, 0.750, and 0.771 in the training set and 0.711, 0.709, 0.684, and 0.674 in the test set, respectively. Regarding the multiple-window setting, the radiomics model based on all four windows showed an AUC of 0.825 in the training set and 0.743 in the test set. Statistically, the 15 models demonstrated comparable performances (P > 0.05).ConclusionA single chest CT window was acceptable in predicting the malignancy of pulmonary nodules, and additional windows did not statistically improve the performance of the radiomics models. In addition, slice thickness and contrast enhancement did not affect the diagnostic performance.

Invited Commentary: Comparing Single vs Double Chest Tube Placement after Decortication for Stage III Empyema.

Journal of the American College of Surgeons ():10.1097/XCS.0000000000000642, March 08, 2023. | DOI: 10.1097/XCS.0000000000000642

Patient and Fetal Radiation-Induced Malignancy Risk From Imaging For Evaluation of Pulmonary Embolism in Pregnancy

BackgroundImaging for diagnosis of suspected pulmonary embolism in pregnancy presents radiation concerns for patient and fetus. ObjectivesEstimate the risks of radiation-induced breast cancer and childhood leukemia from common imaging techniques for the evaluation of suspected pulmonary embolism in pregnancy. MethodsBreast and uterine absorbed doses for various imaging techniques were input into the National Cancer Institute Radiation Risk Assessment Tool to calculate risk of breast cancer for the patient and childhood leukemia for the fetus. Absorbed doses were obtained by synthesizing data from a recent systematic review and the International Commission on Radiological Protection. Primary outcomes were the estimated excess incidences of breast cancer and childhood leukemia per 100,000 exposures. ResultsBaseline incidences of breast cancer for a 30-year-old woman and childhood leukemia for a male fetus were 13,341 and 939, respectively. Excess incidences of breast cancer were 0.003 and 0.275 for a single and two-view chest radiograph, respectively, 9.53 and 20.6 for low- and full-dose computed tomography pulmonary angiography (CTPA), respectively, 0.616 and 2.54 for low- and full-dose perfusion scan, respectively, and 0.732 and 2.66 for low- and full-dose ventilation perfusion scan, respectively. Excess incidences of childhood leukemia were 0.004 and 0.007 for a single and two-view chest radiograph, respectively, 0.069 and 0.490 for low- and full-dose CTPA, respectively, 0.359 and 1.47 for low- and full-dose perfusion scan, respectively, and 0.856 and 1.97 for low- and full-dose ventilation perfusion scan, respectively. ConclusionExcess cancer risks for all techniques were small relative to baseline cancer risks, with CTPA techniques carrying slightly higher risk of breast cancer for the patient and ventilation perfusion techniques a higher risk of childhood leukemia.

Can Imaging Using Radiomics and Fat Fraction Analysis Detect Early Tissue Changes on Historical CT Scans in the Regions of the Pancreas Gland That Subsequently Develop Adenocarcinoma?

Despite a growing number of effective therapeutic options for patients with pancreatic adenocarcinoma, the prognosis remains dismal mostly due to the late-stage presentation and spread of the cancer to other organs. Because a genomic analysis of pancreas tissue revealed that it may take years, if not decades, for pancreatic cancer to develop, we performed radiomics and fat fraction analysis on contrast-enhanced CT (CECT) scans of patients with historical scans showing no evidence of cancer but who subsequently went on to develop pancreas cancer years later, in an attempt to identify specific imaging features of the normal pancreas that may portend the subsequent development of the cancer. In this IRB-exempt, retrospective, single institution study, CECT chest, abdomen, and pelvis (CAP) scans of 22 patients who had evaluable historical imaging data were analyzed. The images from the "healthy" pancreas were obtained between 3.8 and 13.9 years before the diagnosis of pancreas cancer was established. Afterwards, the images were used to divide and draw seven regions of interest (ROIs) around the pancreas (uncinate, head, neck-genu, body (proximal, middle, and distal) and tail). Radiomic analysis on these pancreatic ROIs consisted of first order quantitative texture analysis features such as kurtosis, skewness, and fat quantification. Of all the variables tested, fat fraction in the pancreas tail (p = 0.029) and asymmetry of the histogram frequency curve (skewness) of pancreas tissue (p = 0.038) were identified as the most important imaging signatures for subsequent cancer development. Changes in the texture of the pancreas as measured on the CECT of patients who developed pancreas cancer years later could be identified, confirming the utility of radiomics-based imaging as a potential predictor of oncologic outcomes. Such findings may be potentially useful in the future to screen patients for pancreatic cancer, thereby helping detect pancreas cancer at an early stage and improving survival.

Sybil: A Validated Deep Learning Model to Predict Future Lung Cancer Risk From a Single Low-Dose Chest Computed Tomography.

Low-dose computed tomography (LDCT) for lung cancer screening is effective, although most eligible people are not being screened. Tools that provide personalized future cancer risk assessment could focus approaches toward those most likely to benefit. We hypothesized that a deep learning model assessing the entire volumetric LDCT data could be built to predict individual risk without requiring additional demographic or clinical data. We developed a model called Sybil using LDCTs from the National Lung Screening Trial (NLST). Sybil requires only one LDCT and does not require clinical data or radiologist annotations; it can run in real time in the background on a radiology reading station. Sybil was validated on three independent data sets: a heldout set of 6,282 LDCTs from NLST participants, 8,821 LDCTs from Massachusetts General Hospital (MGH), and 12,280 LDCTs from Chang Gung Memorial Hospital (CGMH, which included people with a range of smoking history including nonsmokers). Sybil achieved area under the receiver-operator curves for lung cancer prediction at 1 year of 0.92 (95% CI, 0.88 to 0.95) on NLST, 0.86 (95% CI, 0.82 to 0.90) on MGH, and 0.94 (95% CI, 0.91 to 1.00) on CGMH external validation sets. Concordance indices over 6 years were 0.75 (95% CI, 0.72 to 0.78), 0.81 (95% CI, 0.77 to 0.85), and 0.80 (95% CI, 0.75 to 0.86) for NLST, MGH, and CGMH, respectively. Sybil can accurately predict an individual's future lung cancer risk from a single LDCT scan to further enable personalized screening. Future study is required to understand Sybil's clinical applications. Our model and annotations are publicly available.