Use of Temporally Validated Machine Learning Models To Predict Outcomes of Percutaneous Nephrolithotomy Using Data from the British Association of Urological Surgeons Percutaneous Nephrolithotomy Audit.

BACKGROUND AND OBJECTIVE: Machine learning (ML) is a subset of artificial intelligence that uses data to build algorithms to predict specific outcomes. Few ML studies have examined percutaneous nephrolithotomy (PCNL) outcomes. Our objective was to build, streamline, temporally validate, and use ML models for prediction of PCNL outcomes (intensive care admission, postoperative infection, transfusion, adjuvant treatment, postoperative complications, visceral injury, and stone-free status at follow-up) using a comprehensive national database (British Association of Urological Surgeons PCNL). METHODS: This was an ML study using data from a prospective national database. Extreme gradient boosting (XGB), deep neural network (DNN), and logistic regression (LR) models were built for each outcome of interest using complete cases only, imputed, and oversampled and imputed/oversampled data sets. All validation was performed with complete cases only. Temporal validation was performed with 2019 data only. A second round used a composite of the most important 11 variables in each model to build the final model for inclusion in the shiny application. We report statistics for prognostic accuracy. KEY FINDINGS AND LIMITATIONS: The database contains 12 810 patients. The final variables included were age, Charlson comorbidity index, preoperative haemoglobin, Guy's stone score, stone location, size of outer sheath, preoperative midstream urine result, primary puncture site, preoperative dimercapto-succinic acid scan, stone size, and image guidance (https://endourology.shinyapps.io/PCNL_Demographics/). The areas under the receiver operating characteristic curve was >0.6 in all cases. CONCLUSIONS AND CLINICAL IMPLICATIONS: This is the largest ML study on PCNL outcomes to date. The models are temporally valid and therefore can be implemented in clinical practice for patient-specific risk profiling. Further work will be conducted to externally validate the models. PATIENT SUMMARY: We applied artificial intelligence to data for patients who underwent a keyhole surgery to remove kidney stones and developed a model to predict outcomes for this procedure. Doctors could use this tool to advise patients about their risk of complications and the outcomes they can expect after this surgery.

Routine Urinary Biochemistry Does Not Accurately Predict Stone Type Nor Recurrence in Kidney Stone Formers: A Multicentre, Multimodel, Externally Validated Machine-Learning Study.

Objectives: Urinary biochemistry is used to detect and monitor conditions associated with recurrent kidney stones. There are no predictive machine learning (ML) tools for kidney stone type or recurrence. We therefore aimed to build and validate ML models for these outcomes using age, gender, 24-hour urine biochemistry, and stone composition. Materials and Methods: Data from three cohorts were used, Southampton, United Kingdom (n = 3013), Newcastle, United Kingdom (n = 5984), and Bern, Switzerland (n = 794). Of these 3130 had available 24-hour urine biochemistry measurements (calcium, oxalate, urate [Ur], pH, volume), and 1684 had clinical data on kidney stone recurrence. Predictive ML models were built for stone type (n = 5 models) and recurrence (n = 7 models) using the UK data, and externally validated with the Swiss data. Three sets of models were built using complete cases, multiple imputation, and oversampling techniques. Results: For kidney stone type one model (extreme gradient boosting [XGBoost] built using oversampled data) was able to effectively discriminate between calcium oxalate, calcium phosphate, and Ur on both internal and external validation. For stone recurrence, none of the models were able to discriminate between recurrent and nonrecurrent stone formers. Conclusions: Kidney stone recurrence cannot be accurately predicted using modeling tools built using specific 24-hour urinary biochemistry values alone. A single model was able to differentiate between stone types. Further studies to delineate accurate predictive tools should be undertaken using both known and novel risk factors, including radiomics and genomics.

Wood-density has no effect on stomatal control of leaf-level water use efficiency in an Amazonian forest.

Forest disturbances increase the proportion of fast-growing tree species compared to slow-growing ones. To understand their relative capacity for carbon uptake and their vulnerability to climate change, and to represent those differences in Earth system models, it is necessary to characterise the physiological differences in their leaf-level control of water use efficiency and carbon assimilation. We used wood density as a proxy for the fast-slow growth spectrum and tested the assumption that trees with a low wood density (LWD) have a lower water-use efficiency than trees with a high wood density (HWD). We selected 5 LWD tree species and 5 HWD tree species growing in the same location in an Amazonian tropical forest and measured in situ steady-state gas exchange on top-of-canopy leaves with parallel sampling and measurement of leaf mass area and leaf nitrogen content. We found that LWD species invested more nitrogen in photosynthetic capacity than HWD species, had higher photosynthetic rates and higher stomatal conductance. However, contrary to expectations, we showed that the stomatal control of the balance between transpiration and carbon assimilation was similar in LWD and HWD species and that they had the same dark respiration rates.

Seasonal trends in leaf-level photosynthetic capacity and water use efficiency in a North American Eastern deciduous forest and their impact on canopy-scale gas exchange.

Vegetative transpiration (E) and photosynthetic carbon assimilation (A) are known to be seasonally dynamic, with changes in their ratio determining the marginal water use efficiency (WUE). Despite an understanding that stomata play a mechanistic role in regulating WUE, it is still unclear how stomatal and nonstomatal processes influence change in WUE over the course of the growing season. As a result, limited understanding of the primary physiological drivers of seasonal dynamics of canopy WUE remains one of the largest uncertainties in earth system model projections of carbon and water exchange in temperate deciduous forest ecosystems. We investigated seasonal patterns in leaf-level physiological, hydraulic, and anatomical properties, including the seasonal progress of the stomatal slope parameter (g1 ; inversely proportional to WUE) and the maximum carboxylation rate (Vcmax ). Vcmax and g1 were seasonally variable; however, their patterns were not temporally synchronized. g1 generally showed an increasing trend until late in the season, while Vcmax peaked during the midsummer months. Seasonal progression of Vcmax was primarily driven by changes in leaf structural, and anatomical characteristics, while seasonal changes in g1 were most strongly related to changes in Vcmax and leaf hydraulics. Using a seasonally variable Vcmax and g1 to parameterize a canopy-scale gas exchange model increased seasonally aggregated A and E by 3% and 16%, respectively.

The effect of the vertical gradients of photosynthetic parameters on the CO2 assimilation and transpiration of a Panamanian tropical forest.

Terrestrial biosphere models (TBMs) include the representation of vertical gradients in leaf traits associated with modeling photosynthesis, respiration, and stomatal conductance. However, model assumptions associated with these gradients have not been tested in complex tropical forest canopies. We compared TBM representation of the vertical gradients of key leaf traits with measurements made in a tropical forest in Panama and then quantified the impact of the observed gradients on simulated canopy-scale CO2 and water fluxes. Comparison between observed and TBM trait gradients showed divergence that impacted canopy-scale simulations of water vapor and CO2 exchange. Notably, the ratio between the dark respiration rate and the maximum carboxylation rate was lower near the ground than at the top-of-canopy, leaf-level water-use efficiency was markedly higher at the top-of-canopy, and the decrease in maximum carboxylation rate from the top-of-canopy to the ground was less than TBM assumptions. The representation of the gradients of leaf traits in TBMs is typically derived from measurements made within-individual plants, or, for some traits, assumed constant due to a lack of experimental data. Our work shows that these assumptions are not representative of the trait gradients observed in species-rich, complex tropical forests.

Infection after ureteroscopy for ureteric stones: analysis of 71 305 cases in the Hospital Episode Statistics database.

OBJECTIVES: To investigate the burden of infectious complications following ureteroscopy (URS) for ureteric stones on a national level in England using data from the Hospital Episodes Statistics (HES) data warehouse. MATERIALS AND METHODS: A retrospective cohort was identified and followed up in HES during the period April 2013 to March 2020 for all procedure codes relating to ureteroscopic stone treatment (M27.1, M27.2, M27.3). Treatment episodes relating to the first URS ('index ureteroscopy') for each patient were further analysed. All subsequent admissions within 30 days were also captured. The primary outcome was diagnosis of urinary tract infection (UTI; including all codes relating to a UTI/sepsis within the first 30 days of index URS). Secondary outcomes were critical care attendance, attendance at the accident and emergency department (A&E) within 30 days, and mortality. RESULTS: A total of 71 305 index ureteroscopies were eligible for analysis. The median age was 55 years, and 81% of procedures were elective and 45% were undertaken as day-cases. At the time of index URS, 16% of patients had diabetes, 0.5% had coexisting neurological disease and 40% had an existing stent/nephrostomy. Overall, 6.8% of the cohort (n = 4822) had a diagnosis of UTI within 30 days of index URS (3.9% immediately after surgery). A total of 339 patients (0.5%) required an unplanned stay in critical care during their index URS admission; 8833 patients (12%) attended A&E within 30 days. Overall mortality was 0.18% (60 in-hospital, 65 within 30 days); 40 deaths (0.056%) included infection as a contributing cause of death. CONCLUSION: We present the largest series evaluating infectious complications after ureteroscopic stone treatment. The procedure is safe, with low inpatient infective complication and critical care admission rates.

Short-term variation in leaf-level water use efficiency in a tropical forest.

The representation of stomatal regulation of transpiration and CO2 assimilation is key to forecasting terrestrial ecosystem responses to global change. Given its importance in determining the relationship between forest productivity and climate, accurate and mechanistic model representation of the relationship between stomatal conductance (gs ) and assimilation is crucial. We assess possible physiological and mechanistic controls on the estimation of the g1 (stomatal slope, inversely proportional to water use efficiency) and g0 (stomatal intercept) parameters, using diurnal gas exchange surveys and leaf-level response curves of six tropical broadleaf evergreen tree species. g1 estimated from ex situ response curves averaged 50% less than g1 estimated from survey data. While g0 and g1 varied between leaves of different phenological stages, the trend was not consistent among species. We identified a diurnal trend associated with g1 and g0 that significantly improved model projections of diurnal trends in transpiration. The accuracy of modeled gs can be improved by accounting for variation in stomatal behavior across diurnal periods, and between measurement approaches, rather than focusing on phenological variation in stomatal behavior. Additional investigation into the primary mechanisms responsible for diurnal variation in g1 will be required to account for this phenomenon in land-surface models.

Leaves as bottlenecks: The contribution of tree leaves to hydraulic resistance within the soil-plant-atmosphere continuum.

Within vascular plants, the partitioning of hydraulic resistance along the soil-to-leaf continuum affects transpiration and its response to environmental conditions. In trees, the fractional contribution of leaf hydraulic resistance (Rleaf ) to total soil-to-leaf hydraulic resistance (Rtotal ), or fRleaf (=Rleaf /Rtotal ), is thought to be large, but this has not been tested comprehensively. We compiled a multibiome data set of fRleaf using new and previously published measurements of pressure differences within trees in situ. Across 80 samples, fRleaf averaged 0.51 (95% confidence interval [CI] = 0.46-0.57) and it declined with tree height. We also used the allometric relationship between field-based measurements of soil-to-leaf hydraulic conductance and laboratory-based measurements of leaf hydraulic conductance to compute the average fRleaf for 19 tree samples, which was 0.40 (95% CI = 0.29-0.56). The in situ technique produces a more accurate descriptor of fRleaf because it accounts for dynamic leaf hydraulic conductance. Both approaches demonstrate the outsized role of leaves in controlling tree hydrodynamics. A larger fRleaf may help stems from loss of hydraulic conductance. Thus, the decline in fRleaf with tree height would contribute to greater drought vulnerability in taller trees and potentially to their observed disproportionate drought mortality.

Role of Renal Biopsy in the Management of Renal Cancer: Concordance between Ultrasound/CT-Guided Biopsy Results and Definitive Pathology, Adverse Events, and Complication Rate.

(1) Background: In the last decade, the number of detected renal cancer cases has increased, with the highest incidence in Western countries. Although renal biopsy is reported as a safe procedure, it is not adopted in all centres. As it is not possible to accurately distinguish benign tumours using imaging, this may lead to overtreatment. Most of the cancer detected on imaging is treated by surgery, radiofrequency ablation (RFA), or cryotherapy. (2) Methods: This was a single-centre retrospective study of 225 patients studied preoperatively with ultrasound (US)/CT-guided renal biopsy, with the aim of supporting clinical management. Decisions regarding the biopsy were based on either MDT indication or physician preference. US-guided renal biopsy was the first option for all patients; CT-guided biopsy was used when US-guided biopsy was not feasible. The efficacy of renal biopsy in terms of diagnostic performance and the concordance between biopsy results and definitive pathology were investigated. Additionally, adverse events related to the biopsy were recorded and analysed. Data collected throughout the study were analysed using binary logistic regression, Fisher's exact test, and Pearson's chi-square test to investigate possible correlations between post-procedural complications and the size of the lesion. (3) Results: Renal biopsy was not diagnostic in 23/225 (10.2%) patients. A CT-guided approach was necessary in 20/225 patients after failure of US-guided biopsy. The complication rate of renal biopsy was 4.8% overall-all Clavien grade I and without any serious sequelae. Interestingly, complications occurred in patients with very different sizes of renal cell carcinoma. No correlation between complications and anticoagulant/antiplatelet drugs was found. No seeding was reported among the patients who underwent partial/radical nephrectomy. (4) Conclusions: Renal biopsy was shown to be safe and effective, with a high concordance between biopsy results and definitive pathology and a low rate of complications. The use of a CT-guided approach whenever the US-guided approach failed improved the diagnostic performance of renal biopsy.

Enabling FAIR data in Earth and environmental science with community-centric (meta)data reporting formats.

Research can be more transparent and collaborative by using Findable, Accessible, Interoperable, and Reusable (FAIR) principles to publish Earth and environmental science data. Reporting formats-instructions, templates, and tools for consistently formatting data within a discipline-can help make data more accessible and reusable. However, the immense diversity of data types across Earth science disciplines makes development and adoption challenging. Here, we describe 11 community reporting formats for a diverse set of Earth science (meta)data including cross-domain metadata (dataset metadata, location metadata, sample metadata), file-formatting guidelines (file-level metadata, CSV files, terrestrial model data archiving), and domain-specific reporting formats for some biological, geochemical, and hydrological data (amplicon abundance tables, leaf-level gas exchange, soil respiration, water and sediment chemistry, sensor-based hydrologic measurements). More broadly, we provide guidelines that communities can use to create new (meta)data formats that integrate with their scientific workflows. Such reporting formats have the potential to accelerate scientific discovery and predictions by making it easier for data contributors to provide (meta)data that are more interoperable and reusable.

Late-day measurement of excised branches results in uncertainty in the estimation of two stomatal parameters derived from response curves in Populus deltoides Bartr. × Populus nigra L.

Many terrestrial biosphere models depend on an understanding of the relationship between stomatal conductance and photosynthesis. However, unlike the measurement of photosynthetic parameters, such as the maximum carboxylation capacity, where standard methods (e.g., CO2 response or ACi curves) are widely accepted, a consensus method for empirically measuring parameters representing stomatal response has not yet emerged. Most models of stomatal response to environment represent stomatal conductance as being bounded by a lower intercept parameter (g0), and linearly scaled based on a multivariate term described by the stomatal slope parameter (g1). Here we employ the widely used Unified Stomatal Optimization model, to test whether g1 and g0 parameters are impacted by the choice of measurement method, either on an intact branch or a cut branch segment stored in water. We measured paired stomatal response curves on intact and excised branches of a hybrid poplar clone (Populus deltoides Bartr. × Populus nigra L. OP367), measured twice over a diurnal period. We found that predawn branch excision did not significantly affect measured g0 and g1 when measured within 4 h of excision. Measurement in the afternoon resulted in significantly higher values of g1 and lower values of g0, with values changing by 55% and 56%, respectively. Excision combined with afternoon measurement resulted in a marked effect on parameter estimates, with g1 increasing 89% from morning to afternoon and a 25% lower g1 for cut branches than those measured in situ. We also show that in hybrid poplar the differences in parameter estimates obtained from plants measured under different conditions can directly impact models of canopy function, reducing modeled transpiration by 18% over a simulated 12.5-h period. Although these results are only for a single isohydric woody species, our findings suggest that stomatal optimality parameters may not remain constant throughout the day.

An improved representation of the relationship between photosynthesis and stomatal conductance leads to more stable estimation of conductance parameters and improves the goodness-of-fit across diverse data sets.

Stomata play a central role in surface-atmosphere exchange by controlling the flux of water and CO2 between the leaf and the atmosphere. Representation of stomatal conductance (gsw ) is therefore an essential component of models that seek to simulate water and CO2 exchange in plants and ecosystems. For given environmental conditions at the leaf surface (CO2 concentration and vapor pressure deficit or relative humidity), models typically assume a linear relationship between gsw and photosynthetic CO2 assimilation (A). However, measurement of leaf-level gsw response curves to changes in A are rare, particularly in the tropics, resulting in only limited data to evaluate this key assumption. Here, we measured the response of gsw and A to irradiance in six tropical species at different leaf phenological stages. We showed that the relationship between gsw and A was not linear, challenging the key assumption upon which optimality theory is based-that the marginal cost of water gain is constant. Our data showed that increasing A resulted in a small increase in gsw at low irradiance, but a much larger increase at high irradiance. We reformulated the popular Unified Stomatal Optimization (USO) model to account for this phenomenon and to enable consistent estimation of the key conductance parameters g0 and g1 . Our modification of the USO model improved the goodness-of-fit and reduced bias, enabling robust estimation of conductance parameters at any irradiance. In addition, our modification revealed previously undetectable relationships between the stomatal slope parameter g1 and other leaf traits. We also observed nonlinear behavior between A and gsw in independent data sets that included data collected from attached and detached leaves, and from plants grown at elevated CO2 concentration. We propose that this empirical modification of the USO model can improve the measurement of gsw parameters and the estimation of plant and ecosystem-scale water and CO2  fluxes.

Post-Ureteroscopy Infections Are Linked to Pre-Operative Stent Dwell Time over Two Months: Outcomes of Three European Endourology Centres.

BACKGROUND: The aim of this study is to investigate outcomes of pre-operative stent dwell time on infectious complications following ureteroscopy and stone treatment to identify a time cut-off. MATERIAL AND METHODS: Three tertiary referral centres in Europe retrospectively collected outcomes of ureteroscopy and laser fragmentation (URSL) for all patients with pre-operative indwelling ureteric stents over a period of up to 5 years. Data was collected on patient details, stone demographics, stent dwell time, complications and stone free rate (SFR). Matching for age, sex, operative time, stone size and post-operative stent insertion. To examine for a threshold effect, monthly cut-offs were used to compare post-ureteroscopic febrile UTIs. Binomial logistic regression was used (SPSS v.24) with a significance level set at 0.0036. The risk ratio (RR) with a 95% confidence interval (CI) and the number needed to harm (NNH) are reported. RESULTS: There were 467 patients with a pre-operative stent for analysis. These patients (n = 315) were matched to non-stented controls after excluding 152 patients to achieve adequate matching. There was a significant difference in rates of post-ureteroscopic febrile UTI between stented vs non-stented patients (RR = 2.67, 95% CI: 1.10-6.48, p = 0.03). On adjustment, a dwell time of more than two months was associated with an increased risk of post-ureteroscopic febrile UTI (RR = 3.94, 95% CI: 1.30-12.01, p = 0.02), this increased risk rose with longer dwell time. At stent time longer than four months was associated with a significantly increased risk of post-ureteroscopic febrile UTI (5% vs. 15%, RR = 3.09, 95% CI: 1.56-6.10, p = 0.001), with the number needed to harm at 10. CONCLUSIONS: Overall infectious complication rates from URSL are low. The risk of post-operative UTI after four months of dwell time is nearly tripled compared to less than four months.

Reducing model uncertainty of climate change impacts on high latitude carbon assimilation.

The Arctic-Boreal Region (ABR) has a large impact on global vegetation-atmosphere interactions and is experiencing markedly greater warming than the rest of the planet, a trend that is projected to continue with anticipated future emissions of CO2 . The ABR is a significant source of uncertainty in estimates of carbon uptake in terrestrial biosphere models such that reducing this uncertainty is critical for more accurately estimating global carbon cycling and understanding the response of the region to global change. Process representation and parameterization associated with gross primary productivity (GPP) drives a large amount of this model uncertainty, particularly within the next 50 years, where the response of existing vegetation to climate change will dominate estimates of GPP for the region. Here we review our current understanding and model representation of GPP in northern latitudes, focusing on vegetation composition, phenology, and physiology, and consider how climate change alters these three components. We highlight challenges in the ABR for predicting GPP, but also focus on the unique opportunities for advancing knowledge and model representation, particularly through the combination of remote sensing and traditional boots-on-the-ground science.

Ten-year survival outcomes after radical nephroureterectomy with a risk-stratified approach using prior diagnostic ureteroscopy: a single-institution observational retrospective cohort study.

OBJECTIVES: To evaluate the long-term oncological outcomes of patients with upper tract urothelial carcinoma (UTUC) undergoing radical nephroureterectomy (RNU) and the impact of diagnostic ureteroscopy (URS) on survival outcomes. MATERIALS AND METHODS: A retrospective analysis of all consecutive patients undergoing RNU for suspected UTUC at a UK tertiary referral centre from a prospectively maintained database was conducted. The primary outcome measures were 5- and 10-year cancer-specific survival (CSS). The secondary outcomes were: overall survival (OS), recurrence-free survival (RFS), impact of prior diagnostic URS on OS, CSS and intravesical RFS (intravesical-RFS), and predictors of intravesical recurrence. Statistical analysis was performed in R using the 'survminer' and 'survival' packages. The Kaplan-Meier method was used to calculate survival functions and these were expressed in graphical form. Uni-/multivariate survival analyses were performed using the Cox proportional hazard regression model. Statistical significance in this study was set at P < 0.05. RESULTS: A total of 422 patients underwent RNU with confirmed UTUC. The median (interquartile range) follow-up of patients with confirmed UTUC was 9.2 (5.6-12.7) years. The 5- and 10-year CSS rates were 70.5% (95% confidence interval [CI] 65.9-74.9) and 67.1% (95% CI 62.4-71.6), respectively. OS (HR 1.04 [95% CI 0.78-1.38]; P = 0.46) and CSS (HR 0.96 [95% CI 0.68-1.34]; P = 0.81) were similar in the diagnostic URS and the direct RNU cohorts. intravesical RFS was superior for the direct RNU cohort (HR 1.94 [95% CI 1.19-3.17]; P = 0.008). In multivariate analysis, prior URS, T2 stage, proximal ureter tumour and bladder cancer history were predictors of metachronous bladder recurrence. CONCLUSION: This single-centre retrospective cohort study reports the long-term oncological outcomes of RNU with a median follow-up of 9.2 years, serving as a reference standard in counselling patients undergoing RNU. Stage and grade of the RNU specimen were the only two studied factors that appeared to adversely impact long-term CSS and OS. Our results suggest that the risk of intravesical recurrence is increased nearly twofold in patients who have undergone diagnostic URS prior to RNU. Prior URS, however, does not appear to adversely impact long-term CSS and OS. The authors suggest that a risk-stratified approach be adopted, wherein diagnostic URS is offered only in equivocal cases.

Rapid estimation of photosynthetic leaf traits of tropical plants in diverse environmental conditions using reflectance spectroscopy.

Tropical forests are one of the main carbon sinks on Earth, but the magnitude of CO2 absorbed by tropical vegetation remains uncertain. Terrestrial biosphere models (TBMs) are commonly used to estimate the CO2 absorbed by forests, but their performance is highly sensitive to the parameterization of processes that control leaf-level CO2 exchange. Direct measurements of leaf respiratory and photosynthetic traits that determine vegetation CO2 fluxes are critical, but traditional approaches are time-consuming. Reflectance spectroscopy can be a viable alternative for the estimation of these traits and, because data collection is markedly quicker than traditional gas exchange, the approach can enable the rapid assembly of large datasets. However, the application of spectroscopy to estimate photosynthetic traits across a wide range of tropical species, leaf ages and light environments has not been extensively studied. Here, we used leaf reflectance spectroscopy together with partial least-squares regression (PLSR) modeling to estimate leaf respiration (Rdark25), the maximum rate of carboxylation by the enzyme Rubisco (Vcmax25), the maximum rate of electron transport (Jmax25), and the triose phosphate utilization rate (Tp25), all normalized to 25°C. We collected data from three tropical forest sites and included leaves from fifty-three species sampled at different leaf phenological stages and different leaf light environments. Our resulting spectra-trait models validated on randomly sampled data showed good predictive performance for Vcmax25, Jmax25, Tp25 and Rdark25 (RMSE of 13, 20, 1.5 and 0.3 µmol m-2 s-1, and R2 of 0.74, 0.73, 0.64 and 0.58, respectively). The models showed similar performance when applied to leaves of species not included in the training dataset, illustrating that the approach is robust for capturing the main axes of trait variation in tropical species. We discuss the utility of the spectra-trait and traditional gas exchange approaches for enhancing tropical plant trait studies and improving the parameterization of TBMs.

A Machine Learning Predictive Model for Post-Ureteroscopy Urosepsis Needing Intensive Care Unit Admission: A Case-Control YAU Endourology Study from Nine European Centres.

INTRODUCTION: With the rise in the use of ureteroscopy and laser stone lithotripsy (URSL), a proportionate increase in the risk of post-procedural urosepsis has also been observed. The aims of our paper were to analyse the predictors for severe urosepsis using a machine learning model (ML) in patients that needed intensive care unit (ICU) admission and to make comparisons with a matched cohort. METHODS: A retrospective study was conducted across nine high-volume endourology European centres for all patients who underwent URSL and subsequently needed ICU admission for urosepsis (Group A). This was matched by patients with URSL without urosepsis (Group B). Statistical analysis was performed with 'R statistical software' using the 'randomforests' package. The data were segregated at random into a 70% training set and a 30% test set using the 'sample' command. A random forests ML model was then built with n = 300 trees, with the test set used for internal validation. Diagnostic accuracy statistics were generated using the 'caret' package. RESULTS: A total of 114 patients were included (57 in each group) with a mean age of 60 ± 16 years and a male:female ratio of 1:1.19. The ML model correctly predicted risk of sepsis in 14/17 (82%) cases (Group A) and predicted those without urosepsis for 12/15 (80%) controls (Group B), whilst overall it also discriminated between the two groups predicting both those with and without sepsis. Our model accuracy was 81.3% (95%, CI: 63.7-92.8%), sensitivity = 0.80, specificity = 0.82 and area under the curve = 0.89. Predictive values most commonly accounting for nodal points in the trees were a large proximal stone location, long stent time, large stone size and long operative time. CONCLUSION: Urosepsis after endourological procedures remains one of the main reasons for ICU admission. Risk factors for urosepsis are reasonably accurately predicted by our innovative ML model. Focusing on these risk factors can allow one to create predictive strategies to minimise post-operative morbidity.

Detection of the metabolic response to drought stress using hyperspectral reflectance.

Drought is the most important limitation on crop yield. Understanding and detecting drought stress in crops is vital for improving water use efficiency through effective breeding and management. Leaf reflectance spectroscopy offers a rapid, non-destructive alternative to traditional techniques for measuring plant traits involved in a drought response. We measured drought stress in six glasshouse-grown agronomic species using physiological, biochemical, and spectral data. In contrast to physiological traits, leaf metabolite concentrations revealed drought stress before it was visible to the naked eye. We used full-spectrum leaf reflectance data to predict metabolite concentrations using partial least-squares regression, with validation R2 values of 0.49-0.87. We show for the first time that spectroscopy may be used for the quantitative estimation of proline and abscisic acid, demonstrating the first use of hyperspectral data to detect a phytohormone. We used linear discriminant analysis and partial least squares discriminant analysis to differentiate between watered plants and those subjected to drought based on measured traits (accuracy: 71%) and raw spectral data (66%). Finally, we validated our glasshouse-developed models in an independent field trial. We demonstrate that spectroscopy can detect drought stress via underlying biochemical changes, before visual differences occur, representing a powerful advance for measuring limitations on yield.