Search

Introduction: Many patients sustaining hip fracture do not regain their pre-injury function. Early rehabilitation improves outcome. The aim of this study was to investigate if hip fracture patients progressed differently following surgery depending on whether arthroplasty or internal fixation (IF) was employed. This information may be useful to identify patient cohorts who may require further rehabilitation prior to discharge home. Methods: A prospective database was used to audit outcomes for hip fracture patients presenting to our unit between October 2019 and October 2020. Our study group comprised 89 patients with femoral neck fractures. 60 patients were female, 29 were male. Average age was 77 years (range 50-96). 69 patients were treated by arthroplasty, 20 patients were treated with IF. The group treated by IF were younger (mean 71.4 years) compared to the group treated by arthroplasty (mean 79.3 years) (P = 0.0027). Timed Up and Go (TUG) test was performed as soon as patients were able to complete same post-operatively, and again at discharge, to assess difference over time. Results: 17 patients out of20 (85%) of the group treated with IF were able to perform a TUG test prior to discharge, at a mean of 4.3 (range 2-13) days post-operatively. Mean time taken to complete an initial TUG by the IF group was 74.7 seconds. Of the patients treated with arthroplasty, 54 out of 69 (78%) were able to complete a TUG prior to discharge, at a mean of 4.9 (range 1-16) days. The mean time taken to complete the initial TUG by a patient treated with arthroplasty was 87.8 seconds. By discharge the time taken to complete this distance had reduced in both groups with the arthroplasty group completing their TUG in a mean of 46 seconds (median 39 seconds) and the IF group completing the TUG in a mean of 55.7 seconds (median 46 seconds). Twenty-six (38%) of the patients treated with arthroplasty were dis-charged home directly, while eleven (55%) of the patients treated with IF were discharged home directly. Conclusion: Hip fracture patients treated with arthroplasty had lower functional ability on day one post-operatively compared to the IF group, however patients treated with arthroplasty showed the greatest improvement in function and mobility at the time of discharge. Despite this, a greater proportion of the IF group were discharged directly home (55% vs 38%) and able to receive any ongoing rehabilitation as an outpatient while the majority of the arthroplasty group required further inpatient rehabilitation. [ABSTRACT FROM AUTHOR]

In 1970, a recovery and exhumation was conducted of 83 mummified bodies interred in an early 19th century Victorian crypt cemetery in Huddersfield, England. Due to the high degree of soft tissue preservation, including skin and organs, forensic pathologists were able to conduct complete autopsies documenting disease states and causes of death. Unique taphonomic conditions of the sealed lead crypts had been produced by the combined factors of: an anaerobic environment, cloth burial shrouds, and oak bark shavings. From a sample of 27 of the best preserved bodies, this study consists of a retrospective review and analysis of the archived postmortem reports, notes and photographs to document the degree and pattern of soft tissue preservation, including adipocere and tanning and their association with types of interment materials and structures. Results of the analysis indicated that 70% of individuals were well preserved, and adipocere was present on 60% of the adults, 33% of children and 50% of the infants. The most common locations for this were the head and face as well as the extremities. The presence of adipocere and the degree of skin tanning were assessed subjectively from the autopsy photographs and ranked on a 5-point scale. This analysis revealed that the presence of oak shavings is positively associated with skin tanning, based on mean tanning scores of 0.77 for individuals without oak and 3.34 out of 5 for individuals with oak, a highly significant result in a Wilcoxon ranked pairs test. Generally, tanning better preserved the skin but not the internal organs, while preservation of all soft tissues was greater when less oak bark and adipocere were present but when burial shrouds were used.

Conformal prediction has seen many applications in pharmaceutical science, being able to calibrate outputs of machine learning models and producing valid prediction intervals. We here present the open source software CPSign that is a complete implementation of conformal prediction for cheminformatics modeling. CPSign implements inductive and transductive conformal prediction for classification and regression, and probabilistic prediction with the Venn-ABERS methodology. The main chemical representation is signatures but other types of descriptors are also supported. The main modeling methodology is support vector machines (SVMs), but additional modeling methods are supported via an extension mechanism, e.g. DeepLearning4J models. We also describe features for visualizing results from conformal models including calibration and efficiency plots, as well as features to publish predictive models as REST services. We compare CPSign against other common cheminformatics modeling approaches including random forest, and a directed message-passing neural network. The results show that CPSign produces robust predictive performance with comparative predictive efficiency, with superior runtime and lower hardware requirements compared to neural network based models. CPSign has been used in several studies and is in production-use in multiple organizations. The ability to work directly with chemical input files, perform descriptor calculation and modeling with SVM in the conformal prediction framework, with a single software package having a low footprint and fast execution time makes CPSign a convenient and yet flexible package for training, deploying, and predicting on chemical data. CPSign can be downloaded from GitHub at https://github.com/arosbio/cpsign .Scientific contribution CPSign provides a single software that allows users to perform data preprocessing, modeling and make predictions directly on chemical structures, using conformal and probabilistic prediction. Building and evaluating new models can be achieved at a high abstraction level, without sacrificing flexibility and predictive performance-showcased with a method evaluation against contemporary modeling approaches, where CPSign performs on par with a state-of-the-art deep learning based model., (© 2024. The Author(s).)

Progressive multiple sclerosis (PMS) is currently diagnosed retrospectively. Here, we work toward a set of biomarkers that could assist in early diagnosis of PMS. A selection of cerebrospinal fluid metabolites (n = 15) was shown to differentiate between PMS and its preceding phenotype in an independent cohort (AUC = 0.93). Complementing the classifier with conformal prediction showed that highly confident predictions could be made, and that three out of eight patients developing PMS within three years of sample collection were predicted as PMS at that time point. Finally, this methodology was applied to PMS patients as part of a clinical trial for intrathecal treatment with rituximab. The methodology showed that 68% of the patients decreased their similarity to the PMS phenotype one year after treatment. In conclusion, the inclusion of confidence predictors contributes with more information compared to traditional machine learning, and this information is relevant for disease monitoring., Competing Interests: The authors declare no competing interests., (© 2023 The Authors.)

Volume of distribution at steady state (V ss ) is an important pharmacokinetic endpoint. In this study we apply machine learning and conformal prediction for human V ss prediction, and make a head-to-head comparison with rat-to-man scaling, allometric scaling and the Rodgers-Lukova method on combined in silico and in vitro data, using a test set of 105 compounds with experimentally observed V ss .The mean prediction error and % with <2-fold prediction error for our method were 2.4-fold and 64%, respectively. 69% of test compounds had an observed V ss within the prediction interval at a 70% confidence level. In comparison, 2.2-, 2.9- and 3.1-fold mean errors and 69, 64 and 61% of predictions with <2-fold error was reached with rat-to-man and allometric scaling and Rodgers-Lukova method, respectively.We conclude that our method has theoretically proven validity that was empirically confirmed, and showing predictive accuracy on par with animal models and superior to an alternative widely used in silico -based method. The option for the user to select the level of confidence in predictions offers better guidance on how to optimise V ss in drug discovery applications.

Machine learning is widely used in drug development to predict activity in biological assays based on chemical structure. However, the process of transitioning from one experimental setup to another for the same biological endpoint has not been extensively studied. In a retrospective study, we here explore different modeling strategies of how to combine data from the old and new assays when training conformal prediction models using data from hERG and Na V assays. We suggest to continuously monitor the validity and efficiency of models as more data is accumulated from the new assay and select a modeling strategy based on these metrics. In order to maximize the utility of data from the old assay, we propose a strategy that augments the proper training set of an inductive conformal predictor by adding data from the old assay but only having data from the new assay in the calibration set, which results in valid (well-calibrated) models with improved efficiency compared to other strategies. We study the results for varying sizes of new and old assays, allowing for discussion of different practical scenarios. We also conclude that our proposed assay transition strategy is more beneficial, and the value of data from the new assay is higher, for the harder case of regression compared to classification problems.

Machine learning methods are widely used in drug discovery and toxicity prediction. While showing overall good performance in cross-validation studies, their predictive power (often) drops in cases where the query samples have drifted from the training data's descriptor space. Thus, the assumption for applying machine learning algorithms, that training and test data stem from the same distribution, might not always be fulfilled. In this work, conformal prediction is used to assess the calibration of the models. Deviations from the expected error may indicate that training and test data originate from different distributions. Exemplified on the Tox21 datasets, composed of chronologically released Tox21Train, Tox21Test and Tox21Score subsets, we observed that while internally valid models could be trained using cross-validation on Tox21Train, predictions on the external Tox21Score data resulted in higher error rates than expected. To improve the prediction on the external sets, a strategy exchanging the calibration set with more recent data, such as Tox21Test, has successfully been introduced. We conclude that conformal prediction can be used to diagnose data drifts and other issues related to model calibration. The proposed improvement strategy-exchanging the calibration data only-is convenient as it does not require retraining of the underlying model.

One of the challenges with predictive modeling is how to quantify the reliability of the models' predictions on new objects. In this work we give an introduction to conformal prediction, a framework that sits on top of traditional machine learning algorithms and which outputs valid confidence estimates to predictions from QSAR models in the form of prediction intervals that are specific to each predicted object. For regression, a prediction interval consists of an upper and a lower bound. For classification, a prediction interval is a set that contains none, one, or many of the potential classes. The size of the prediction interval is affected by a user-specified confidence/significance level, and by the nonconformity of the predicted object; i.e., the strangeness as defined by a nonconformity function. Conformal prediction provides a rigorous and mathematically proven framework for in silico modeling with guarantees on error rates as well as a consistent handling of the models' applicability domain intrinsically linked to the underlying machine learning model. Apart from introducing the concepts and types of conformal prediction, we also provide an example application for modeling ABC transporters using conformal prediction, as well as a discussion on general implications for drug discovery., (Copyright © 2020 The Authors. Published by Elsevier Inc. All rights reserved.)

People seeking asylum (PSA) are recommended to undertake a comprehensive risk-based health assessment within 1 month of arrival in Australia. The Asylum Seeker Resource Centre (ASRC) offers health services to PSA in Victoria, through the ASRC nurse-led clinic. A healthcare assessment is conducted by nurses using a Refugee Health Assessment (RHA) tool. A process evaluation was conducted to assess if the adapted 2016 version of the RHA tool was able to appropriately identify, describe and prioritise the needs of PSA. Twelve ASRC nurses who conducted assessments were interviewed. All interviews were audio recorded, transcribed and analysed. The adapted RHA tool was considered as appropriate for identifying, describing and prioritising the needs of PSA. Three key interconnected themes emerged influencing the appropriateness of the tool: the tool; users' experiences; and the individual characteristics of the PSA undergoing the assessment. Key tool limitations included: the structure not being user-friendly; variability in users' sensitivity to using the tool; and the limited feedback from PSA on the adapted RHA tool. Given the high number of people seeking asylum in Australia and the complexity of their healthcare needs, this research provides preliminary guidance on ways to improve the appropriateness of the ASRC-adapted RHA tool.

Immunization uptake rates are assuming considerable importance as performance indicators for district health authorities (DHAs) and general practitioners (GPs). Data from the Cover of Vaccination Evaluated Rapidly (COVER) programme suggest that immunization uptake rates in innercity districts are well below the 90 per cent target. Using Parkside DHA, a district with consistently poor uptake, as an example, this paper has examined the accuracy of the Child Health Computer (CHC) records on which the COVER figures are based. In October 1989 a cohort of 1485 children born between April and June 1988 were identified. The 462 children identified by the CHC as immunization defaulters (using diphtheria and tetanus immunizations) were followed up by contacting child health clinics, GP surgeries and parents. This exercise revealed that 195 children were in fact immunized and 176 children were erroneously listed, largely because they were no longer resident at the address given. Only a small proportion of this information had reached the CHC by the end of the follow-up period. In addition to high population mobility, information inaccuracies appeared to be caused partly by a failure in data transfer between GPs and the DHA and by failures in data transfer between districts. Overall, the results of the study suggest that there is a considerable discrepancy between actual and recorded immunization uptake levels in Parkside and probably other inner-city areas. This will have a considerable impact on GPs and other health professionals who are striving to achieve 90 per cent targets.

A new method of collecting and analysing sweat (Wescor Macroduct) has advantages of simplicity of collection and direct reading of results by osmometry. Forty-seven children with cystic fibrosis and 47 normal children had sweat tests performed simultaneously by the Gibson and Cooke method and by the Wescor Macroduct method. The new method had a higher rate of inadequate collection (19% vs 6%) which was more marked in children under 5 years of age. This was due partly to the difficulty of fitting a 2.5 cm pilocarpine gel disc to small arms. When an adequate collection was obtained, results were reliable with no false negatives occurring during this study. The Wescor Macroduct sweat test is a reliable method for use in peripheral centres to screen patients suspected of having cystic fibrosis. All children with an inadequate collection or a positive result should be referred to a reference centre for confirmation of the diagnosis. However, the majority will be saved the expense and disruption of travel.

Previous studies on human cortical area 39 suggested that neuron:glial ratios differed between the sexes. These findings were the inspiration for the present investigation which dealt with neuronal and glial counts in area 39 in the male and female rat cerebral cortex. Transverse, celloidin or frozen sections, were cut from male and female brains (respectively) from 90-day-old Long-Evans rats. Neurons and glia were counted on enlarged photographs of stained sections, including area 39, with 35-mm Kodak Panatomic-X film using a Zeiss photomicroscope (X400). Five-by-three-inch prints were taped together in sequence to yield a 640X enlarged "montage" of area 39. Five cell types were differentiated with reference to a standard: neurons, astrocytes, oligodendrocytes, "dark astrocytes," and unidentified glia. The data were analyzed with a two-way analysis of variance (ANOVA: five cell types by two hemispheres). Student's t test and a paired t test were used when appropriate. The neuron:glial ratios in the male rats were consistently higher than those in the females in both hemispheres. The male right side had 12% (P less than 0.05) more neurons than the left; the female had 13% (P less than 0.05) more neurons on the left than the right. Similar, but not identical, asymmetrical patterns were seen with the glial cells.