Investigating the Unfavored Factors That Interfere MALDI-TOF Based AI in Predicting Antibiotic Resistance

Combining Matrix-Assisted Laser Desorption/Ionization Time-of-Flight (MALDI-TOF) spectra data and artificial intelligence (AI) has been introduced for rapid prediction on antibiotic susceptibility test (AST) of S. aureus. Based on the AI predictive probability, the cases with probabilities between low and high cut-offs are defined as “grey zone”. We aimed to investigate the underlying reasons of unconfident (grey zone) or wrong predictive AST. A total 479 S. aureus isolates were collected, analyzed by MALDI-TOF, and AST prediction, standard AST were obtained in a tertiary medical center. The predictions were categorized into the correct prediction group, wrong prediction group, and grey zone group. We analyzed the association between the predictive results and the demographic data, spectral data, and strain types. For MRSA, larger cefoxitin zone size was found in the wrong prediction group. MLST of the MRSA isolates in the grey zone group revealed that uncommon strain types composed 80%. Amid MSSA isolates in the grey zone group, the majority (60%) was composed of over 10 different strain types. In predicting AST based on MALDI-TOF AI, uncommon strains and high diversity would contribute to suboptimal predictive performance.

Hypothermia for Patients Requiring Evacuation of Subdural Hematoma: A Multicenter Randomized Clinical Trial

Background Hypothermia is neuroprotective in some ischemia–reperfusion injuries. Ischemia–reperfusion injury may occur with traumatic subdural hematoma (SDH). This study aimed to determine whether early induction and maintenance of hypothermia in patients with acute SDH would lead to decreased ischemia–reperfusion injury and improve global neurologic outcome. Methods This international, multicenter randomized controlled trial enrolled adult patients with SDH requiring evacuation of hematoma within 6 h of injury. The intervention was controlled temperature management of hypothermia to 35 °C prior to dura opening followed by 33 °C for 48 h compared with normothermia (37 °C). Investigators randomly assigned patients at a 1:1 ratio between hypothermia and normothermia. Blinded evaluators assessed outcome using a 6-month Glasgow Outcome Scale Extended score. Investigators measured circulating glial fibrillary acidic protein and ubiquitin C-terminal hydrolase L1 levels. Results Independent statisticians performed an interim analysis of 31 patients to assess the predictive probability of success and the Data and Safety Monitoring Board recommended the early termination of the study because of futility. Thirty-two patients, 16 per arm, were analyzed. Favorable 6-month Glasgow Outcome Scale Extended outcomes were not statistically significantly different between hypothermia vs. normothermia groups (6 of 16, 38% vs. 4 of 16, 25%; odds ratio 1.8 [95% confidence interval 0.39 to ∞], p = .35). Plasma levels of glial fibrillary acidic protein (p = .036), but not ubiquitin C-terminal hydrolase L1 (p = .26), were lower in the patients with favorable outcome compared with those with unfavorable outcome, but differences were not identified by temperature group. Adverse events were similar between groups. Conclusions This trial of hypothermia after acute SDH evacuation was terminated because of a low predictive probability of meeting the study objectives. There was no statistically significant difference in functional outcome identified between temperature groups.

Fifty Ways to Tag your Pubtypes: Multi-Tagger, a Set of Probabilistic Publication Type and Study Design Taggers to Support Biomedical Indexing and Evidence-Based Medicine

Objective: Indexing articles according to publication types (PTs) and study designs can be a great aid to filtering literature for information retrieval, especially for evidence syntheses. In this study, 50 automated machine learning based probabilistic PT and study design taggers were built and applied to all articles in PubMed. Materials and Methods: PubMed article metadata from 1987-2014 were used as training data, with 2015 used for recalibration. The set of articles indexed with a particular study design MeSH term or PT tag was used as positive training sets. For each PT, the rest of the literature from the same time period was used as its negative training set. Multiple features based on each article title, abstract and metadata were used in training the models. Taggers were evaluated on PubMed articles from 2016 and 2019. A manual analysis was also performed. Results: Of the 50 predictive models that we created, 44 of these achieved an AUC of ~0.90 or greater, with many having performance above 0.95. Of the clinically related study designs, the best performing was SYSTEMATIC_REVIEW with an AUC of 0.998; the lowest performing was RANDOM_ALLOCATION, with an AUC of 0.823. Discussion: This work demonstrates that is feasible to build a large set of probabilistic publication type and study design taggers with high accuracy and ranking performance. Automated tagging permits users to identify qualifying articles as soon as they are published, and allows consistent criteria to be applied across different bibliographic databases. Probabilistic predictive scores are more flexible than binary yes/no predictions, since thresholds can be tailored for specific uses such as high recall literature search, user-adjustable retrieval size, and quality improvement of manually annotated databases. Conclusion: The PT predictive probability scores for all PubMed articles are freely downloadable at http://arrowsmith.psych.uic.edu/evidence_based_medicine/mt_download.html for incorporation into user tools and workflows. Users can also perform PubMed queries at our Anne O'Tate value-added PubMed search engine http://arrowsmith.psych.uic.edu/cgi-bin/arrowsmith_uic/AnneOTate.cgi and filter retrieved articles according to both NLM-annotated and model-predicted publication types and study designs.

Relationship between the difference in electric pulp test values and the diagnostic type of pulpitis

Background According to the diagnosis criteria of the American Association of Endodontists (AAE), sensitive responses to cold and/or heat tests of suspected teeth compared with those of control teeth can be used for the diagnosis of pulpitis, but the role of electric pulp test (EPT) is not mentioned. It is believed that EPT has some limitations in determining the vitality of the pulp. The aim of this study was to explore the association between the difference in EPT values and the differential diagnoses of reversible pulpitis (RP) and symptomatic irreversible pulpitis (SIRP) caused by dental caries. Methods A total of 203 cases with pulpitis caused by dental caries were included. A diagnosis of pulpitis was made on the basis of the diagnostic criteria of AAE. Patient demographic and clinical examination data were collected. The EPT values of the suspected teeth and control teeth were measured, and the differences between them were calculated. The correlation between the difference in the EPT values and diagnosis of pulpitis was analyzed using univariate and multivariate logistic regression. Results In the 203 cases (78 males and 125 females; 115 cases of RP, 88 cases of SIRP; 9 anterior teeth, 59 premolars, and 135 molars), the mean patient age was 34.04 ± 13.02 (standard deviation) years. The unadjusted (crude) model, model 1 (adjusted for age), model 2 (adjusted for age and sex), and model 3 (adjusted for age, sex, and tooth type) were established for the statistical analyses. In model 3 [odds ratio (OR) = 1.025; 95% confidence interval (CI) 1.002–1.050; P = 0.035], the difference in EPT values between RP and SIRP was statistically significant. However, the areas under the curve of predictive probability of the crude model, model 1, model 2, and model 3 were 0.565, 0.570, 0.585, and 0.617, respectively, showing that the model accuracy was low. The P-value for the trend in differences between the EPT values as a categorical variable showed that the differences in the EPT values, comparing RP and SIRP, were not statistically significant. Conclusions Based on the present data, the difference in EPT values was not sufficient to differentiate RP from SIRP.

Illustrative Discussion of MC-Dropout in General Dataset: Uncertainty Estimation in Bitcoin

The past few years have witnessed the resurgence of uncertainty estimation generally in neural networks. Providing uncertainty quantification besides the predictive probability is desirable to reflect the degree of belief in the model’s decision about a given input. Recently, Monte-Carlo dropout (MC-dropout) method has been introduced as a probabilistic approach based Bayesian approximation which is computationally efficient than Bayesian neural networks. MC-dropout has revealed promising results on image datasets regarding uncertainty quantification. However, this method has been subjected to criticism regarding the behaviour of MC-dropout and what type of uncertainty it actually captures. For this purpose, we aim to discuss the behaviour of MC-dropout on classification tasks using synthetic and real data. We empirically explain different cases of MC-dropout that reflects the relative merits of this method. Our main finding is that MC-dropout captures datapoints lying on the decision boundary between the opposed classes using synthetic data. On the other hand, we apply MC-dropout method on dataset derived from Bitcoin known as Elliptic data to highlight the outperformance of model with MC-dropout over standard model. A conclusion and possible future directions are proposed.