Can machine learning predict pharmacotherapy outcomes? An application study in osteoporosis

Development and Validation of a Machine Learning-Based Nomogram for Prediction of Unplanned Reoperation Postspinal Surgery Within 30 Days

BACKGROUND

Unplanned reoperation postspinal surgery (URPS) leads to prolonged hospital stays, higher costs, decreased patient satisfaction, and adversely affects postoperative rehabilitation. This study aimed to develop and validate prediction models (nomograms) for early URPS risk factors using machine learning methods, aiding spine surgeons in designing prevention strategies, promoting early recovery, reducing complications, and improving patient satisfaction.

METHODS

Medical records of 639 patients who underwent reoperation postspinal surgery from the First Affiliated Hospital of Air Force Medical University (2018-2022) were collected, including baseline indicators, perioperative indicators, and laboratory indicators. After applying inclusion and exclusion criteria, 122 URPS and 155 non-URPS patients were identified and randomly divided into training (82 URPS and 111 non-URPS) and validation (40 URPS and 44 non-URPS) cohorts. Three machine learning methods (least absolute shrinkage and selection operator regression, Random Forest, and Support Vector Machine Recursive Feature Elimination) were used to select feature variables, and their intersection was used to develop the prediction model, tested on the validation cohort.

RESULTS

Six factors-implant, postoperative suction drainage, gelatin sponge, anticoagulants, antibiotics, and disease type-were identified to construct a nomogram diagnostic model. The area under the curve of this nomogram was 0.829 (95% confidence interval 0.771-0.886) in the training cohort and 0.854 (95% confidence interval 0.775-0.933) in the validation cohort. Calibration curves demonstrated satisfactory agreement between predictions and actual probabilities. The decision curve indicated clinical usefulness with a threshold between 1% and 90%.

CONCLUSIONS

The established model can effectively predict URPS in patients and can assist spine surgeons in devising personalized and rational clinical prevention strategies.

Diagnosis Osteoporosis Risk: Using Machine Learning Algorithms Among Fasa Adults Cohort Study (FACS)

INTRODUCTION

In Iran, the assessment of osteoporosis through tools like dual-energy X-ray absorptiometry poses significant challenges due to their high costs and limited availability, particularly in small cities and rural areas. Our objective was to employ a variety of machine learning (ML) techniques to evaluate the accuracy and precision of each method, with the aim of identifying the most accurate pattern for diagnosing the osteoporosis risks.

METHODS

We analysed the data related to osteoporosis risk factors obtained from the Fasa Adults Cohort Study in eight ML methods, including logistic regression (LR), baseline LR, decision tree classifiers (DT), support vector classifiers (SVC), random forest classifiers (RF), linear discriminant analysis (LDA), K nearest neighbour classifiers (KNN) and extreme gradient boosting (XGB). For each algorithm, we calculated accuracy, precision, sensitivity, specificity, F1 score and area under the curve (AUC) and compared them.

RESULTS

The XGB model with an AUC of 0.78 (95% confidence interval [CI]: 0.74-0.82) and an accuracy of 0.79 (0.75-0.83) demonstrated the best performance, while AUC and accuracy values of RF were achieved (0.78 and 0.77), LR (0.78 and 0.77), LDA (0.78 and 0.76), DT (0.76 and 0.79), SVC (0.71 and 0.64), KNN (0.63 and 0.59) and baseline LR (0.72 and 0.82), respectively.

CONCLUSION

The XGB model had the best performance in assessing the risk of osteoporosis in the Iranian population. Given the disadvantages and challenges associated with traditional osteoporosis diagnostic tools, the implementation of ML algorithms for the early identification of individuals with osteoporosis can lead to a significant reduction in morbidity and mortality related to this condition. This advancement not only alleviates the substantial financial burden placed on the healthcare systems of various countries due to the treatment of complications arising from osteoporosis but also encourages health policies to shift toward more preventive approaches for managing this disease.

Combining clinical and molecular data for personalized treatment in acute myeloid leukemia: A machine learning approach

BACKGROUND AND OBJECTIVE

The standard of care in Acute Myeloid Leukemia patients has remained essentially unchanged for nearly 40 years. Due to the complicated mutational patterns within and between individual patients and a lack of targeted agents for most mutational events, implementing individualized treatment for AML has proven difficult. We reanalysed the BeatAML dataset employing Machine Learning algorithms. The BeatAML project entails patients extensively characterized at the molecular and clinical levels and linked to drug sensitivity outputs. Our approach capitalizes on the molecular and clinical data provided by the BeatAML dataset to predict the ex vivo drug sensitivity for the 122 drugs evaluated by the project.

METHODS

We utilized ElasticNet, which produces fully interpretable models, in combination with a two-step training protocol that allowed us to narrow down computations. We automated the genes' filtering step by employing two metrics, and we evaluated all possible data combinations to identify the best training configuration settings per drug.

RESULTS

We report a Pearson correlation across all drugs of 0.36 when clinical and RNA sequencing data were combined, with the best-performing models reaching a Pearson correlation of 0.67. When we trained using the datasets in isolation, we noted that RNA Sequencing data (Pearson: 0.36) attained three times the predictive power of whole exome sequencing data (Pearson: 0.11), with clinical data falling somewhere in between (Pearson 0.26). Lastly, we present a paradigm of clinical significance. We used our models' prediction as a drug sensitivity score to rank an individual's expected response to treatment. We identified 78 patients out of 89 (88 %) that the proposed drug was more potent than the administered one based on their ex vivo drug sensitivity data.

CONCLUSIONS

In conclusion, our reanalysis of the BeatAML dataset using Machine Learning algorithms demonstrates the potential for individualized treatment prediction in Acute Myeloid Leukemia patients, addressing the longstanding challenge of treatment personalization in this disease. By leveraging molecular and clinical data, our approach yields promising correlations between predicted drug sensitivity and actual responses, highlighting a significant step forward in improving therapeutic outcomes for AML patients.

Predicting and refining acid modifications of biochar based on machine learning and bibliometric analysis: Specific surface area, average pore size, and total pore volume

Acid-modified biochar is a modified biochar material with convenient preparation, high specific surface area, and rich pore structure. It has great potential for application in the heavy metal remediation, soil amendments, and carrying catalysts. Specific surface area (SSA), average pore size (APS), and total pore volume (TPV) are the key properties that determine its adsorption capacity, reactivity, and water holding capacity, and an intensive study of these properties is essential to optimize the performance of biochar. But the complex interactions among the preparation conditions obstruct finding the optimal modification strategy. This study collected dataset through bibliometric analysis and used four typical machine learning models to predict the SSA, APS, and TPV of acid-modified biochar. The results showed that the extreme gradient boosting (XGB) was optimal for the test results (SSA R2 = 0.92, APS R2 = 0.87, TPV R2 = 0.96). The model interpretation revealed that the modification conditions were the major factors affecting SSA and TPV, and the pyrolysis conditions were the major factors affecting APS. Based on the XGB model, the modification conditions of biochar were optimized, which revealed the ideal preparation conditions for producing the optimal biochar (SSA = 727.02 m2/g, APS = 5.34 nm, TPV = 0.68 cm3/g). Moreover, the biochar produced under specific conditions verified the generalization ability of the XGB model (R2 = 0.99, RMSE = 12.355). This study provides guidance for optimizing the preparation strategy of acid-modified biochar and promotes its potentiality for industrial application.

Modelling future bone mineral density: Simplicity or complexity?

BACKGROUND

Osteoporotic fractures are a major global public health issue, leading to patient suffering and death, and considerable healthcare costs. Bone mineral density (BMD) measurement is important to identify those with osteoporosis and assess their risk of fracture. Both the absolute BMD and the change in BMD over time contribute to fracture risk. Predicting future fracture in individual patients is challenging and impacts clinical decisions such as when to intervene or repeat BMD measurement. Although the importance of BMD change is recognised, an effective way to incorporate this marginal effect into clinical algorithms is lacking.

METHODS

We compared two methods using longitudinal DXA data generated from subjects with two or more hip DXA scans on the same machine between 2000 and 2018. A simpler statistical method (ZBM) was used to predict an individual's future BMD based on the mean BMD and the standard deviation of the reference group and their BMD measured in the latest scan. A more complex deep learning (DL)-based method was developed to cope with multidimensional longitudinal data, variables extracted from patients' historical DXA scan(s), as well as features drawn from the ZBM method. Sensitivity analyses of several subgroups was conducted to evaluate the performance of the derived models.

RESULTS

2948 white adults aged 40-90 years met our study inclusion: 2652 (90 %) females and 296 (10 %) males. Our DL-based models performed significantly better than the ZBM models in women, particularly our Hybrid-DL model. In contrast, the ZBM-based models performed as well or better than DL-based models in men.

CONCLUSIONS

Deep learning-based and statistical models have potential to forecast future BMD using longitudinal clinical data. These methods have the potential to augment clinical decisions regarding when to repeat BMD testing in the assessment of osteoporosis.

Prediction of subsequent fragility fractures: application of machine learning

BACKGROUND

Machine learning (ML) has shown exceptional promise in various domains of medical research. However, its application in predicting subsequent fragility fractures is still largely unknown. In this study, we aim to evaluate the predictive power of different ML algorithms in this area and identify key features associated with the risk of subsequent fragility fractures in osteoporotic patients.

METHODS

We retrospectively analyzed data from patients presented with fragility fractures at our Fracture Liaison Service, categorizing them into index fragility fracture (n = 905) and subsequent fragility fracture groups (n = 195). We independently trained ML models using 27 features for both male and female cohorts. The algorithms tested include Random Forest, XGBoost, CatBoost, Logistic Regression, LightGBM, AdaBoost, Multi-Layer Perceptron, and Support Vector Machine. Model performance was evaluated through 10-fold cross-validation.

RESULTS

The CatBoost model outperformed other models, achieving 87% accuracy and an AUC of 0.951 for females, and 93.4% accuracy with an AUC of 0.990 for males. The most significant predictors for females included age, serum C-reactive protein (CRP), 25(OH)D, creatinine, blood urea nitrogen (BUN), parathyroid hormone (PTH), femoral neck Z-score, menopause age, number of pregnancies, phosphorus, calcium, and body mass index (BMI); for males, the predictors were serum CRP, femoral neck T-score, PTH, hip T-score, BMI, BUN, creatinine, alkaline phosphatase, and spinal Z-score.

CONCLUSION

ML models, especially CatBoost, offer a valuable approach for predicting subsequent fragility fractures in osteoporotic patients. These models hold the potential to enhance clinical decision-making by supporting the development of personalized preventative strategies.

Applications of Artificial Intelligence Methods for the Prediction of Osteoporotic Fractures

Background: Osteoporosis is a socio-economic problem of modern aging societies. Bone fractures and the related treatments generate the highest costs. The occurrence of osteoporotic fractures is a cause of chronic disability, many complications, reduced quality of life, and often premature death. Aim of the study: The aim of the study was to determine which of the patient's potential risk factors pertaining to various diseases and lifestyle have an essential impact on the occurrence of low-energy fractures and the hierarchy of these factors. Methods: The study was retrospective. The documentation of 222 patients (206 women and 16 men) from an osteoporosis treatment clinic in Łódź, Poland was analyzed. Each patient was described by a vector consisting of 27 features, where each feature was a different risk factor. Using artificial neural networks, an attempt was made to create a model that, based on the available data, would be able to predict whether the patient would be exposed to low-energy fractures. We developed a neural network model that achieved the best result for the testing data. In addition, we used other methods to solve the classification problem, i.e., correctly dividing patients into two groups: those with fractures and those without fractures. These methods were logistic regression, k-nearest neighbors and SVM. Results: The obtained results gave us the opportunity to assess the effectiveness of various methods and the importance of the features describing patients. Using logistic regression and the recursive elimination of features, a ranking of risk factors was obtained in which the most important were age, chronic kidney disease, neck T-score, and serum phosphate level. Then, we repeated the learning procedure of the neural network considering only these four most important features. The average mean squared error on the test set was about 27% for the best variant of the model. Conclusions: The comparison of the rankings with different numbers of patients shows that the applied method is very sensitive to changes in the considered data (adding new patients significantly changes the result). Further cohort studies with more patients and more advanced methods of machine learning may be needed to identify other significant risk factors and to develop a reliable fracture risk system. The obtained results may contribute to the improved identification patients at risk of low-energy fractures and early implementation of comprehensive treatment.

A cohort of patients in New York State with an alcohol use disorder and subsequent treatment information - A merging of two administrative data sources

OBJECTIVE

Despite the high prevalence of alcohol use disorder (AUD) in the United States, limited research is focused on the associations among AUD, pain, and opioids/benzodiazepine use. In addition, little is known regarding individuals with a history of AUD and their potential risk for pain diagnoses, pain prescriptions, and subsequent misuse. Moreover, the potential risk of pain diagnoses, prescriptions, and subsequent misuse among individuals with a history of AUD is not well known. The objective was to develop a tailored dataset by linking data from 2 New York State (NYS) administrative databases to investigate a series of hypotheses related to AUD and painful medical disorders.

METHODS

Data from the NYS Office of Addiction Services and Supports (OASAS) Client Data System (CDS) and Medicaid claims data from the NYS Department of Health Medicaid Data Warehouse (MDW) were merged using a stepwise deterministic method. Multiple patient-level identifier combinations were applied to create linkage rules. We included patients aged 18 and older from the OASAS CDS who initially entered treatment with a primary substance use of alcohol and no use of opioids between January 1, 2003, and September 23, 2019. This cohort was then linked to corresponding Medicaid claims.

RESULTS

A total of 177,685 individuals with a primary AUD problem and no opioid use history were included in the dataset. Of these, 37,346 (21.0%) patients had an OUD diagnosis, and 3,365 (1.9%) patients experienced an opioid overdose. There were 121,865 (68.6%) patients found to have a pain condition.

CONCLUSION

The integrated database allows researchers to examine the associations among AUD, pain, and opioids/benzodiazepine use, and propose hypotheses to improve outcomes for at-risk patients. The findings of this study can contribute to the development of a prognostic prediction model and the analysis of longitudinal outcomes to improve the care of patients with AUD.

Predicting the Severity of Adverse Events on Osteoporosis Drugs Using Attribute Weighted Logistic Regression

Osteoporosis is a serious bone disease that affects many people worldwide. Various drugs have been used to treat osteoporosis. However, these drugs may cause severe adverse events in patients. Adverse drug events are harmful reactions caused by drug usage and remain one of the leading causes of death in many countries. Predicting serious adverse drug reactions in the early stages can help save patients' lives and reduce healthcare costs. Classification methods are commonly used to predict the severity of adverse events. These methods usually assume independence among attributes, which may not be practical in real-world applications. In this paper, a new attribute weighted logistic regression is proposed to predict the severity of adverse drug events. Our method relaxes the assumption of independence among the attributes. An evaluation was performed on osteoporosis data obtained from the United States Food and Drug Administration databases. The results showed that our method achieved a higher recognition performance and outperformed baseline methods in predicting the severity of adverse drug events.

Potential of vibrational spectroscopy coupled with machine learning as a non-invasive diagnostic method for COVID-19

BACKGROUND AND OBJECTIVE

Efforts to alleviate the ongoing coronavirus disease 2019 (COVID-19) crisis showed that rapid, sensitive, and large-scale screening is critical for controlling the current infection and that of ongoing pandemics.

METHODS

Here, we explored the potential of vibrational spectroscopy coupled with machine learning to screen COVID-19 patients in its initial stage. Herein presented is a hybrid classification model called grey wolf optimized support vector machine (GWO-SVM). The proposed model was tested and comprehensively compared with other machine learning models via vibrational spectroscopic fingerprinting including saliva FTIR spectra dataset and serum Raman scattering spectra dataset.

RESULTS

For the unknown vibrational spectra, the presented GWO-SVM model provided an accuracy, specificity and F1_score value of 0.9825, 0.9714 and 0.9778 for saliva FTIR spectra dataset, respectively, while an overall accuracy, specificity and F1_score value of 0.9085, 0.9552 and 0.9036 for serum Raman scattering spectra dataset, respectively, which showed superiority than those of state-of-the-art models, thereby suggesting the suitability of the GWO-SVM model to be adopted in a clinical setting for initial screening of COVID-19 patients.

CONCLUSIONS

Prospectively, the presented vibrational spectroscopy based GWO-SVM model can facilitate in screening of COVID-19 patients and alleviate the medical service burden. Therefore, herein proof-of-concept results showed the chance of vibrational spectroscopy coupled with GWO-SVM model to help COVID-19 diagnosis and have the potential be further used for early screening of other infectious diseases.

Medical optimization of osteoporosis for adult spinal deformity surgery: a state-of-the-art evidence-based review of current pharmacotherapy

PURPOSE

Osteoporosis is a common, but challenging phenomenon to overcome in adult spinal deformity (ASD) surgery. Several pharmacological agents are at the surgeon's disposal to optimize the osteoporotic patient prior to undergoing extensive reconstruction. Familiarity with these medications will allow the surgeon to make informed decisions on selecting the most appropriate adjuncts for each individual patient.

METHODS

A comprehensive literature review was conducted in PubMed from September 2021 to April 2022. Studies were selected that contained combinations of various terms including osteoporosis, specific medications, spine surgery, fusion, cage subsidence, screw loosening, pull-out, junctional kyphosis/failure.

RESULTS

Bisphosphonates, denosumab, selective estrogen receptor modulators, teriparatide, abaloparatide and romosozumab are all pharmacological agents currently available for adjunctive use. While these medications have been shown to have beneficial effects on improving bone mineral density in the osteoporotic patient, varying evidence is available on their specific effects in the context of extensive spine surgery. There is still a lack of human studies with use of the newer agents.

CONCLUSION

Bisphosphonates are first-line agents due to their low cost and robust evidence behind their utility. However, in the absence of contraindications, optimizing bone quality with anabolic medications should be strongly considered in preparation for spinal deformity surgeries due to their beneficial and favorable effects on fusion and hardware compared to the anti-resorptive medications.