The relational modeling of hierarchical data in biodiversity databases.

The unifying element of all biodiversity data is the issue of taxon hierarchy modeling. We compared 25 existing databases in terms of handling taxa hierarchy and presentation of this data. We used documentation or demo installations of databases as a source of information and next in line was the analysis of structures using R packages provided by inspected platforms. If neither of these was available, we used the public interface of individual databases. For almost half (12) of the databases analyzed, we did not find any formalized taxa hierarchy data structure, providing only biological information about taxon membership in higher ranks, which is not fully formalizable and thus not generally usable. The least effective Adjacency List model (storing parentId of a taxon) dominates among the remaining providers. This study demonstrates the lack of attention paid by current biodiversity databases to modeling taxon hierarchy, particularly to making it available to researchers in the form of a hierarchical data structure within the data provided. For biodiversity relational databases, the Closure Table type is the most suitable of the known data models, which also corresponds to the ontology concept. However, its use is rather sporadic within the biodiversity databases ecosystem.

Long-term impact of COVID-19 hospitalisation among individuals with pre-existing airway diseases in the UK: a multicentre, longitudinal cohort study - PHOSP-COVID.

Background: The long-term outcomes of COVID-19 hospitalisation in individuals with pre-existing airway diseases are unknown. Methods: Adult participants hospitalised for confirmed or clinically suspected COVID-19 and discharged between 5 March 2020 and 31 March 2021 were recruited to the Post-hospitalisation COVID-19 (PHOSP-COVID) study. Participants attended research visits at 5 months and 1 year post discharge. Clinical characteristics, perceived recovery, burden of symptoms and health-related quality of life (HRQoL) of individuals with pre-existing airway disease (i.e., asthma, COPD or bronchiectasis) were compared to the non-airways group. Results: A total of 615 out of 2697 (22.8%) participants had a history of pre-existing airway diseases (72.0% diagnosed with asthma, 22.9% COPD and 5.1% bronchiectasis). At 1 year, the airways group participants were less likely to feel fully recovered (20.4% versus 33.2%, p<0.001), had higher burden of anxiety (29.1% versus 22.0%, p=0.002), depression (31.2% versus 24.7%, p=0.006), higher percentage of impaired mobility using short physical performance battery ≤10 (57.4% versus 45.2%, p<0.001) and 27% had a new disability (assessed by the Washington Group Short Set on Functioning) versus 16.6%, p=0.014. HRQoL assessed using EQ-5D-5L Utility Index was lower in the airways group (mean±SD 0.64±0.27 versus 0.73±0.25, p<0.001). Burden of breathlessness, fatigue and cough measured using a study-specific tool was higher in the airways group. Conclusion: Individuals with pre-existing airway diseases hospitalised due to COVID-19 were less likely to feel fully recovered, had lower physiological performance measurements, more burden of symptoms and reduced HRQoL up to 1 year post-hospital discharge.

A Fast HPLC/UV Method for Determination of Ketoprofen in Cellular Media.

A simple, sensitive and quick HPLC method was developed for the determination of ketoprofen in cell culture media (EMEM, DMEM, RPMI). Separation was performed using a gradient on the C18 column with a mobile phase of acetonitrile and miliQ water acidified by 0.1 % (v/v) formic acid. The method was validated for parameters including linearity, accuracy, precision, limit of quantitation and limit of detection, as well as robustness. The response was found linear over the range of 3-100 µg/mL as demonstrated by the acquired value of correlation coefficient R2=0.9997. The described method is applicable for determination of various pharmacokinetic aspects of ketoprofen in vitro.

Identifying key features of digital resources used during online science practicals.

As in our everyday lives, we use digital elements as part of formal and informal education. To serve their educational purpose well, systematic research is desirable to identify and measure their characteristics. This study focuses on science practicals, which are complex and vary in organizational settings and specific arrangements, including usage of digital elements. We describe the digital resources on which the online instruction of science practicals during the COVID-19 forced lockdowns was built, and their key characteristics were identified. Data were collected from science teachers in Slovakia, Czechia, Slovenia, France, and Spain. The teachers shared the web resources they used and that they would recommend, together with a description of the resources. We recorded 89 inputs representing 50 unique resources. Teachers preferred free resources, mostly for knowledge revision, and newly discovered 36% of them due to forced distant teaching. The best evaluated resources were those supporting interaction (especially among peers), focused on teaching subjects and/or ICT, ready to use, and with a clear structure. The resource most frequently mentioned and used in more than half of the countries was PhET (Interactive Simulations for Science and Math) which provides free simulations of scientific principles. Other characteristics mentioned in the literature (e.g., supporting creativity and independent solving, connecting different levels of organization, authenticity, flexibility) were not that important for the overall rating.

TET protein inhibitors: Potential and limitations.

TET proteins (methylcytosine dioxygenases) play an important role in the regulation of gene expression. Dysregulation of their activity is associated with many serious pathogenic states such as oncological diseases. Regulation of their activity by specific inhibitors could represent a promising therapeutic strategy. Therefore, this review describes various types of TET protein inhibitors in terms of their inhibitory mechanism and possible applicability. The potential and possible limitations of this approach are thoroughly discussed in the context of TET protein functionality in living systems. Furthermore, possible therapeutic strategies based on the inhibition of TET proteins are presented and evaluated, especially in the field of oncological diseases.

Respiratory sequelae of COVID-19: pulmonary and extrapulmonary origins, and approaches to clinical care and rehabilitation.

Although the exact prevalence of post-COVID-19 condition (also known as long COVID) is unknown, more than a third of patients with COVID-19 develop symptoms that persist for more than 3 months after SARS-CoV-2 infection. These sequelae are highly heterogeneous in nature and adversely affect multiple biological systems, although breathlessness is a frequently cited symptom. Specific pulmonary sequelae, including pulmonary fibrosis and thromboembolic disease, need careful assessment and might require particular investigations and treatments. COVID-19 outcomes in people with pre-existing respiratory conditions vary according to the nature and severity of the respiratory disease and how well it is controlled. Extrapulmonary complications such as reduced exercise tolerance and frailty might contribute to breathlessness in post-COVID-19 condition. Non-pharmacological therapeutic options, including adapted pulmonary rehabilitation programmes and physiotherapy techniques for breathing management, might help to attenuate breathlessness in people with post-COVID-19 condition. Further research is needed to understand the origins and course of respiratory symptoms and to develop effective therapeutic and rehabilitative strategies.

Targeting of the Mitochondrial TET1 Protein by Pyrrolo[3,2-b]pyrrole Chelators.

Targeting of epigenetic mechanisms, such as the hydroxymethylation of DNA, has been intensively studied, with respect to the treatment of many serious pathologies, including oncological disorders. Recent studies demonstrated that promising therapeutic strategies could potentially be based on the inhibition of the TET1 protein (ten-eleven translocation methylcytosine dioxygenase 1) by specific iron chelators. Therefore, in the present work, we prepared a series of pyrrolopyrrole derivatives with hydrazide (1) or hydrazone (2-6) iron-binding groups. As a result, we determined that the basic pyrrolo[3,2-b]pyrrole derivative 1 was a strong inhibitor of the TET1 protein (IC50 = 1.33 µM), supported by microscale thermophoresis and molecular docking. Pyrrolo[3,2-b]pyrroles 2-6, bearing substituted 2-hydroxybenzylidene moieties, displayed no significant inhibitory activity. In addition, in vitro studies demonstrated that derivative 1 exhibits potent anticancer activity and an exclusive mitochondrial localization, confirmed by Pearson's correlation coefficient of 0.92.

Non-psychotropic cannabinoids as inhibitors of TET1 protein.

Non-psychotropic cannabinoids (e.g., cannabidiol, cannabinol and cannabigerol) are contained in numerous alimentary and medicinal products. Therefore, predicting and studying their possible side effects, such as changes in DNA methylation, is an important task for assessing the safety of these products. Interference with TET enzymes by chelating ferrous ions can contribute to the altered methylation pattern. All tested cannabinoids displayed a strong affinity for Fe(II) ions. Cannabidiol and cannabinol exhibited potent inhibitory activities (IC50 = 4.8 and 6.27 µM, respectively) towards the TET1 protein, whereas cannabigerol had no effect on the enzyme activity. An in silico molecular docking study revealed marked binding potential within the catalytic cavity for CBD/CBN, but some affinity was also found for CBG, thus the total lack of activity remains unexplained. These results imply that cannabinoids could affect the activity of the TET1 protein not only due to their affinity for Fe(II) but also due to other types of interactions (e.g., hydrophobic interactions and hydrogen bonding).

Pladias platform: Technical description of the database structure.

Background: Digitising and aggregating local floristic data is a critical step in the study of biodiversity. The integrative web-based platform Pladias, designed to cover a wide range of data on vascular plants, was recently developed in the Czech Republic. The combination of occurrence data with species characteristics opens many opportunities for data analysis and synthesis. New information: This article describes the relational structure of the Pladias database service (PladiasDB) and the context of the platform architecture. The structure is relatively complex, as our goal was to cover: (i) species occurrence records, including their management, validation and export of revised species distribution maps, (ii) data on species characteristics with quality control tools using defined data types and (iii) separate user interfaces (UI) for professionals and the general public. We discuss the approaches chosen to model individual elements in PladiasDB and summarise the experience gained during the first five years of operation of the Pladias platform.

Astegolimab, an anti-ST2, in chronic obstructive pulmonary disease (COPD-ST2OP): a phase 2a, placebo-controlled trial.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a heterogeneous inflammatory airway disease. The epithelial-derived IL-33 and its receptor ST2 have been implicated in airway inflammation and infection. We aimed to determine whether astegolimab, a selective ST2 IgG2 monoclonal antibody, reduces exacerbations in COPD. METHODS: COPD-ST2OP was a single-centre, randomised, double-blinded, placebo-controlled phase 2a trial in moderate-to-very severe COPD. Participants were randomly assigned (1:1) with a web-based system to received 490 mg subcutaneous astegolimab or subcutaneous placebo, every 4 weeks for 44 weeks. The primary endpoint was exacerbation rate assessed for 48 weeks assessed with a negative binomial count model in the intention-to-treat population, with prespecified subgroup analysis by baseline blood eosinophil count. The model was the number of exacerbations over the 48-week treatment period, with treatment group as a covariate. Safety was assessed in the whole study population until week 60. Secondary endpoints included Saint George's Respiratory Questionnaire for COPD (SGRQ-C), FEV1, and blood and sputum cell counts. The trial was registered with ClinicalTrials.gov, NCT03615040. FINDINGS: The exacerbation rate at 48 weeks in the intention-to-treat analysis was not significantly different between the astegolimab group (2·18 [95% CI 1·59 to 2·78]) and the placebo group (2·81 [2·05 to 3·58]; rate ratio 0·78 [95% CI 0·53 to 1·14]; p=0·19]). In the prespecified analysis stratifying patients by blood eosinophil count, patients with 170 or fewer cells per µL had 0·69 exacerbations (0·39 to 1·21), whereas those with more than 170 cells per µL had 0·83 exacerbations (0·49 to 1·40). For the secondary outcomes, the mean difference between the SGRQ-C in the astegolimab group versus placebo group was -3·3 (95% CI -6·4 to -0·2; p=0·039), and mean difference in FEV1 between the two groups was 40 mL (-10 to 90; p=0·094). The difference in geometric mean ratios between the two groups for blood eosinophil counts was 0·59 (95% CI 0·51 to 0·69; p<0·001) and 0·25 (0·19 to 0·33; p<0·001) for sputum eosinophil counts. Incidence of treatment-emergent adverse events was similar between groups. INTERPRETATION: In patients with moderate-to-very severe COPD, astegolimab did not significantly reduce exacerbation rate, but did improve health status compared with placebo. FUNDING: Funded by Genentech and National Institute for Health Research Biomedical Research Centres.

Low Concentrated Fractionalized Nanofibers as Suitable Fillers for Optimization of Structural-Functional Parameters of Dead Space Gel Implants after Rectal Extirpation.

Dead space after rectal resection in colorectal surgery is an area with a high risk of complications. In this study, our goal was to develop a novel 3D implant based on composite hydrogels enriched with fractionalized nanofibers. We employed, as a novel approach in abdominal surgery, the application of agarose gels functionalized with fractionalized nanofibers on pieces dozens of microns large with a well-preserved nano-substructure. This retained excellent cell accommodation and proliferation, while nanofiber structures in separated islets allowed cells a free migration throughout the gel. We found these low-concentrated fractionalized nanofibers to be a good tool for structural and biomechanical optimization of the 3D hydrogel implants. In addition, this nano-structuralized system can serve as a convenient drug delivery system for a controlled release of encapsulated bioactive substances from the nanofiber core. Thus, we present novel 3D nanofiber-based gels for controlled release, with a possibility to modify both their biomechanical properties and drug release intended for 3D lesions healing after a rectal extirpation, hysterectomy, or pelvic exenteration.

Software infrastructure and data pipelines established for technical interoperability within a cross-border cooperation for the flora of the Bohemian Forest.

Background: The timely and geographical resolutions, as well as the quantity and taxon concepts of records on the occurrence of plants near national borders is often ambiguous. This is due to the regional focus and different approaches of the contributing national and regional databases and networks of the neighbouring countries. Careful data transformation between national data providers is essential for understanding distribution patterns and its dynamics for organisms in areas along the national borders. Sharing occurrence data through the international data aggregator Global Biodiversity Information Facility (GBIF) is also complicated and has to consider that the underlying taxonomic concept and geographic information system of each single GBIF dataset might be different. In addition, some regional data providers have a restrictive (non-cc) licensing policy which does not allow data publication via the GBIF network. Therefore, it is necessary to investigate new ways to make data fit for use for a better and comprehensive understanding of the Flora of the Bohemian Forest. New information: In this paper, we present a bilateral technical interoperability solution for vascular plant occurrence data for the area between the Czech Republic and Bavaria. We describe the initial state of data providers in both countries and the factual and technical challenges in finding a sustainable concept to establish mutual data sharing. The resulting solution for a functional infrastructure and an agreed data pipeline is described in a step-by-step approach. The new distributed infrastructure allows botanists and other stakeholders from both countries to work within the cross-border context of historical and current plants' distribution.

The long-term sequelae of COVID-19: an international consensus on research priorities for patients with pre-existing and new-onset airways disease.

Persistent ill health after acute COVID-19-referred to as long COVID, the post-acute COVID-19 syndrome, or the post-COVID-19 condition-has emerged as a major concern. We undertook an international consensus exercise to identify research priorities with the aim of understanding the long-term effects of acute COVID-19, with a focus on people with pre-existing airways disease and the occurrence of new-onset airways disease and associated symptoms. 202 international experts were invited to submit a minimum of three research ideas. After a two-phase internal review process, a final list of 98 research topics was scored by 48 experts. Patients with pre-existing or post-COVID-19 airways disease contributed to the exercise by weighting selected criteria. The highest-ranked research idea focused on investigation of the relationship between prognostic scores at hospital admission and morbidity at 3 months and 12 months after hospital discharge in patients with and without pre-existing airways disease. High priority was also assigned to comparisons of the prevalence and severity of post-COVID-19 fatigue, sarcopenia, anxiety, depression, and risk of future cardiovascular complications in patients with and without pre-existing airways disease. Our approach has enabled development of a set of priorities that could inform future research studies and funding decisions. This prioritisation process could also be adapted to other, non-respiratory aspects of long COVID.

Lung cancer prediction by Deep Learning to identify benign lung nodules.

INTRODUCTION: Deep Learning has been proposed as promising tool to classify malignant nodules. Our aim was to retrospectively validate our Lung Cancer Prediction Convolutional Neural Network (LCP-CNN), which was trained on US screening data, on an independent dataset of indeterminate nodules in an European multicentre trial, to rule out benign nodules maintaining a high lung cancer sensitivity. METHODS: The LCP-CNN has been trained to generate a malignancy score for each nodule using CT data from the U.S. National Lung Screening Trial (NLST), and validated on CT scans containing 2106 nodules (205 lung cancers) detected in patients from from the Early Lung Cancer Diagnosis Using Artificial Intelligence and Big Data (LUCINDA) study, recruited from three tertiary referral centers in the UK, Germany and Netherlands. We pre-defined a benign nodule rule-out test, to identify benign nodules whilst maintaining a high sensitivity, by calculating thresholds on the malignancy score that achieve at least 99 % sensitivity on the NLST data. Overall performance per validation site was evaluated using Area-Under-the-ROC-Curve analysis (AUC). RESULTS: The overall AUC across the European centers was 94.5 % (95 %CI 92.6-96.1). With a high sensitivity of 99.0 %, malignancy could be ruled out in 22.1 % of the nodules, enabling 18.5 % of the patients to avoid follow-up scans. The two false-negative results both represented small typical carcinoids. CONCLUSION: The LCP-CNN, trained on participants with lung nodules from the US NLST dataset, showed excellent performance on identification of benign lung nodules in a multi-center external dataset, ruling out malignancy with high accuracy in about one fifth of the patients with 5-15 mm nodules.

Fuel in Markov Decision Processes (FiMDP): A Practical Approach to Consumption.

Consumption Markov Decision Processes (CMDPs) are probabilistic decision-making models of resource-constrained systems. We introduce FiMDP, a tool for controller synthesis in CMDPs with LTL objectives expressible by deterministic Büchi automata. The tool implements the recent algorithm for polynomial-time controller synthesis in CMDPs, but extends it with many additional features. On the conceptual level, the tool implements heuristics for improving the expected reachability times of accepting states, and a support for multi-agent task allocation. On the practical level, the tool offers (among other features) a new strategy simulation framework, integration with the Storm model checker, and FiMDPEnv - a new set of CMDPs that model real-world resource-constrained systems. We also present an evaluation of FiMDP on these real-world scenarios.

Evaluation of a novel deep learning-based classifier for perifissural nodules.

OBJECTIVES: To evaluate the performance of a novel convolutional neural network (CNN) for the classification of typical perifissural nodules (PFN). METHODS: Chest CT data from two centers in the UK and The Netherlands (1668 unique nodules, 1260 individuals) were collected. Pulmonary nodules were classified into subtypes, including "typical PFNs" on-site, and were reviewed by a central clinician. The dataset was divided into a training/cross-validation set of 1557 nodules (1103 individuals) and a test set of 196 nodules (158 individuals). For the test set, three radiologically trained readers classified the nodules into three nodule categories: typical PFN, atypical PFN, and non-PFN. The consensus of the three readers was used as reference to evaluate the performance of the PFN-CNN. Typical PFNs were considered as positive results, and atypical PFNs and non-PFNs were grouped as negative results. PFN-CNN performance was evaluated using the ROC curve, confusion matrix, and Cohen's kappa. RESULTS: Internal validation yielded a mean AUC of 91.9% (95% CI 90.6-92.9) with 78.7% sensitivity and 90.4% specificity. For the test set, the reader consensus rated 45/196 (23%) of nodules as typical PFN. The classifier-reader agreement (k = 0.62-0.75) was similar to the inter-reader agreement (k = 0.64-0.79). Area under the ROC curve was 95.8% (95% CI 93.3-98.4), with a sensitivity of 95.6% (95% CI 84.9-99.5), and specificity of 88.1% (95% CI 81.8-92.8). CONCLUSION: The PFN-CNN showed excellent performance in classifying typical PFNs. Its agreement with radiologically trained readers is within the range of inter-reader agreement. Thus, the CNN-based system has potential in clinical and screening settings to rule out perifissural nodules and increase reader efficiency. KEY POINTS: • Agreement between the PFN-CNN and radiologically trained readers is within the range of inter-reader agreement. • The CNN model for the classification of typical PFNs achieved an AUC of 95.8% (95% CI 93.3-98.4) with 95.6% (95% CI 84.9-99.5) sensitivity and 88.1% (95% CI 81.8-92.8) specificity compared to the consensus of three readers.

Work Ability among Upper-Secondary School Teachers: Examining the Role of Burnout, Sense of Coherence, and Work-Related and Lifestyle Factors.

Maintaining and promoting teachers' work ability is essential for increasing productivity and preventing early exit from the teaching profession. This study aimed to identify the predictors of work ability among upper-secondary school teachers and examine the mediating role of burnout. A large and diverse group of Czech upper-secondary school teachers was surveyed to address this goal. The sample comprised 531 upper-secondary school teachers (50.0 ± 9.94 years, 19.9 ± 10.62 in the teaching profession, 57.6% females). Relatively greater empirical support was found for the effects of burnout, sense of coherence, work-life balance, and perceived relationships in the school environment on work ability than for the impact of age, homeroom teacher duties, workload, and caring for elderly relatives. Furthermore, burnout served as an important mediator of the relationship between sense of coherence and work ability. Teachers with a higher sense of coherence are thus better able to cope with adverse work circumstances and identify and mobilize internal and external resources to prevent professional exhaustion and the subsequent decline in work ability. The study can guide interventions on the work ability of teachers.

Plant Nitrilase Homologues in Fungi: Phylogenetic and Functional Analysis with Focus on Nitrilases in Trametes versicolor and Agaricus bisporus.

Fungi contain many plant-nitrilase (NLase) homologues according to database searches. In this study, enzymes NitTv1 from Trametes versicolor and NitAb from Agaricus bisporus were purified and characterized as the representatives of this type of fungal NLase. Both enzymes were slightly more similar to NIT4 type than to NIT1/NIT2/NIT3 type of plant NLases in terms of their amino acid sequences. Expression of the synthetic genes in Escherichia coli Origami B (DE3) was induced with 0.02 mM isopropyl ß-D-1-thiogalactopyranoside at 20 °C. Purification of NitTv1 and NitAb by cobalt affinity chromatography gave ca. 6.6 mg and 9.6 mg of protein per 100 mL of culture medium, respectively. Their activities were determined with 25 mM of nitriles in 50 mM Tris/HCl buffer, pH 8.0, at 30 °C. NitTv1 and NitAb transformed ß-cyano-L-alanine (ß-CA) with the highest specific activities (ca. 132 and 40 U mg-1, respectively) similar to plant NLase NIT4. ß-CA was transformed into Asn and Asp as in NIT4 but at lower Asn:Asp ratios. The fungal NLases also exhibited significant activities for (aryl)aliphatic nitriles such as 3-phenylpropionitrile, cinnamonitrile and fumaronitrile (substrates of NLase NIT1). NitTv1 was more stable than NitAb (at pH 5-9 vs. pH 5-7). These NLases may participate in plant-fungus interactions by detoxifying plant nitriles and/or producing plant hormones. Their homology models elucidated the molecular interactions with various nitriles in their active sites.

Assessing the Accuracy of a Deep Learning Method to Risk Stratify Indeterminate Pulmonary Nodules.

Rationale: The management of indeterminate pulmonary nodules (IPNs) remains challenging, resulting in invasive procedures and delays in diagnosis and treatment. Strategies to decrease the rate of unnecessary invasive procedures and optimize surveillance regimens are needed.Objectives: To develop and validate a deep learning method to improve the management of IPNs.Methods: A Lung Cancer Prediction Convolutional Neural Network model was trained using computed tomography images of IPNs from the National Lung Screening Trial, internally validated, and externally tested on cohorts from two academic institutions.Measurements and Main Results: The areas under the receiver operating characteristic curve in the external validation cohorts were 83.5% (95% confidence interval [CI], 75.4-90.7%) and 91.9% (95% CI, 88.7-94.7%), compared with 78.1% (95% CI, 68.7-86.4%) and 81.9 (95% CI, 76.1-87.1%), respectively, for a commonly used clinical risk model for incidental nodules. Using 5% and 65% malignancy thresholds defining low- and high-risk categories, the overall net reclassifications in the validation cohorts for cancers and benign nodules compared with the Mayo model were 0.34 (Vanderbilt) and 0.30 (Oxford) as a rule-in test, and 0.33 (Vanderbilt) and 0.58 (Oxford) as a rule-out test. Compared with traditional risk prediction models, the Lung Cancer Prediction Convolutional Neural Network was associated with improved accuracy in predicting the likelihood of disease at each threshold of management and in our external validation cohorts.Conclusions: This study demonstrates that this deep learning algorithm can correctly reclassify IPNs into low- or high-risk categories in more than a third of cancers and benign nodules when compared with conventional risk models, potentially reducing the number of unnecessary invasive procedures and delays in diagnosis.

External validation of a convolutional neural network artificial intelligence tool to predict malignancy in pulmonary nodules.

BACKGROUND: Estimation of the risk of malignancy in pulmonary nodules detected by CT is central in clinical management. The use of artificial intelligence (AI) offers an opportunity to improve risk prediction. Here we compare the performance of an AI algorithm, the lung cancer prediction convolutional neural network (LCP-CNN), with that of the Brock University model, recommended in UK guidelines. METHODS: A dataset of incidentally detected pulmonary nodules measuring 5-15 mm was collected retrospectively from three UK hospitals for use in a validation study. Ground truth diagnosis for each nodule was based on histology (required for any cancer), resolution, stability or (for pulmonary lymph nodes only) expert opinion. There were 1397 nodules in 1187 patients, of which 234 nodules in 229 (19.3%) patients were cancer. Model discrimination and performance statistics at predefined score thresholds were compared between the Brock model and the LCP-CNN. RESULTS: The area under the curve for LCP-CNN was 89.6% (95% CI 87.6 to 91.5), compared with 86.8% (95% CI 84.3 to 89.1) for the Brock model (p≤0.005). Using the LCP-CNN, we found that 24.5% of nodules scored below the lowest cancer nodule score, compared with 10.9% using the Brock score. Using the predefined thresholds, we found that the LCP-CNN gave one false negative (0.4% of cancers), whereas the Brock model gave six (2.5%), while specificity statistics were similar between the two models. CONCLUSION: The LCP-CNN score has better discrimination and allows a larger proportion of benign nodules to be identified without missing cancers than the Brock model. This has the potential to substantially reduce the proportion of surveillance CT scans required and thus save significant resources.