Comparison of Methods to Reduce Bias From Clinical Prediction Models of Postpartum Depression.

Importance: The lack of standards in methods to reduce bias for clinical algorithms presents various challenges in providing reliable predictions and in addressing health disparities. Objective: To evaluate approaches for reducing bias in machine learning models using a real-world clinical scenario. Design, Setting, and Participants: Health data for this cohort study were obtained from the IBM MarketScan Medicaid Database. Eligibility criteria were as follows: (1) Female individuals aged 12 to 55 years with a live birth record identified by delivery-related codes from January 1, 2014, through December 31, 2018; (2) greater than 80% enrollment through pregnancy to 60 days post partum; and (3) evidence of coverage for depression screening and mental health services. Statistical analysis was performed in 2020. Exposures: Binarized race (Black individuals and White individuals). Main Outcomes and Measures: Machine learning models (logistic regression [LR], random forest, and extreme gradient boosting) were trained for 2 binary outcomes: postpartum depression (PPD) and postpartum mental health service utilization. Risk-adjusted generalized linear models were used for each outcome to assess potential disparity in the cohort associated with binarized race (Black or White). Methods for reducing bias, including reweighing, Prejudice Remover, and removing race from the models, were examined by analyzing changes in fairness metrics compared with the base models. Baseline characteristics of female individuals at the top-predicted risk decile were compared for systematic differences. Fairness metrics of disparate impact (DI, 1 indicates fairness) and equal opportunity difference (EOD, 0 indicates fairness). Results: Among 573 634 female individuals initially examined for this study, 314 903 were White (54.9%), 217 899 were Black (38.0%), and the mean (SD) age was 26.1 (5.5) years. The risk-adjusted odds ratio comparing White participants with Black participants was 2.06 (95% CI, 2.02-2.10) for clinically recognized PPD and 1.37 (95% CI, 1.33-1.40) for postpartum mental health service utilization. Taking the LR model for PPD prediction as an example, reweighing reduced bias as measured by improved DI and EOD metrics from 0.31 and -0.19 to 0.79 and 0.02, respectively. Removing race from the models had inferior performance for reducing bias compared with the other methods (PPD: DI = 0.61; EOD = -0.05; mental health service utilization: DI = 0.63; EOD = -0.04). Conclusions and Relevance: Clinical prediction models trained on potentially biased data may produce unfair outcomes on the basis of the chosen metrics. This study's results suggest that the performance varied depending on the model, outcome label, and method for reducing bias. This approach toward evaluating algorithmic bias can be used as an example for the growing number of researchers who wish to examine and address bias in their data and models.

Precision Medicine, AI, and the Future of Personalized Health Care.

The convergence of artificial intelligence (AI) and precision medicine promises to revolutionize health care. Precision medicine methods identify phenotypes of patients with less-common responses to treatment or unique healthcare needs. AI leverages sophisticated computation and inference to generate insights, enables the system to reason and learn, and empowers clinician decision making through augmented intelligence. Recent literature suggests that translational research exploring this convergence will help solve the most difficult challenges facing precision medicine, especially those in which nongenomic and genomic determinants, combined with information from patient symptoms, clinical history, and lifestyles, will facilitate personalized diagnosis and prognostication.

Patient-Specific Modeling for Structural Heart Intervention: Role of 3D Printing Today and Tomorrow CME.

Structural heart interventions (SHIs) are increasingly applicable in a wide range of heart defects, but the intricate and dynamic nature of cardiac structures can make SHIs challenging to perform. Three-dimensional (3D) printed modeling integrates advanced clinical imaging and 3D printing technology to replicate patient-specific anatomy for comprehensive planning and simulation of SHIs. This review discusses the basic principles of patient-specific 3D print model development, print material selection, and model fabrication and highlights how cardiovascular 3D printing can be used in preprocedural planning, device sizing, enhanced communication, and procedure simulation.

Opportunities for multiscale computational modelling of serotonergic drug effects in Alzheimer's disease.

Alzheimer's disease (AD) is an age-specific neurodegenerative disease that compromises cognitive functioning and impacts the quality of life of an individual. Pathologically, AD is characterised by abnormal accumulation of beta-amyloid (Aß) and hyperphosphorylated tau protein. Despite research advances over the last few decades, there is currently still no cure for AD. Although, medications are available to control some behavioural symptoms and slow the disease's progression, most prescribed medications are based on cholinesterase inhibitors. Over the last decade, there has been increased attention towards novel drugs, targeting alternative neurotransmitter pathways, particularly those targeting serotonergic (5-HT) system. In this review, we focused on 5-HT receptor (5-HTR) mediated signalling and drugs that target these receptors. These pathways regulate key proteins and kinases such as GSK-3 that are associated with abnormal levels of Aß and tau in AD. We then review computational studies related to 5-HT signalling pathways with the potential for providing deeper understanding of AD pathologies. In particular, we suggest that multiscale and multilevel modelling approaches could potentially provide new insights into AD mechanisms, and towards discovering novel 5-HTR based therapeutic targets.

The three-dimensional printing renaissance of individualized anatomical modeling: Are we repeating history?

Three-dimensional (3D) printing has revolutionized individualized medicine for patient-specific anatomical modeling and surgical planning. The surge of investigations in model creation for preoperative assessments and patient education has demonstrated improvements in both operative factors and patient satisfaction. In addition, recent technologic advances in 3D printing techniques have provided a resource to create visually pleasing models with chromatic cues for segmentation of adjacent structures. Despite these advances, an important consideration that has yet to be addressed is the quality of representation of the not only the form of structures created, but also the functional relationships of each structure. Jean François Fernel (1497-1558 AD) recognized a similar trend in anatomic innovation over 500 years ago, and sparked a series of texts that challenged the superficial anthropocentric views of the time and led to the foundation of physiologic principles that shaped modern medical philosophy. Accurately generating anatomical structures are directly related to discerning true physiologic function, and a comprehensive understanding of both is essential to hold accountability in fidelity for individualized 3D printing.

Oncología personalizada: principales biomarcadores en el pronóstico y tratamiento de tumores sólidos / Personalised oncology: main biomarkers in the prognosis and treatment of solid tumours

En las últimas décadas ha habido grandes avances en los tratamientos personalizados en pacientes oncológicos gracias a un importante desarrollo científico. El análisis genómico ha mostrado que tumores que parecían tener un origen común, en realidad constituyen un grupo de diversas entidades moleculares. Por otro lado, ha sido muy importante el desarrollo de fármacos dirigidos que actúan de forma específica en las rutas bioquímicas involucradas en el proceso oncológico. El conocimiento de la biología celular y molecular del cáncer ha hecho posible identificar los mecanismos responsables de la transformación maligna y está permitiendo utilizar nuevos marcadores de especial utilidad para definir el pronóstico y determinar el tratamiento de las enfermedades oncológicas

Due to significant scientific developments in the last decades, treatment for oncology patients has started to use more specific and personalised approaches. The genomic analysis has demonstrated that tumours that seemed similar constitute a diverse group of molecular entities. One of the most important breakthroughs is the development of targeted drugs aimed at specific biochemical pathways. Recent advances in knowledge of the cellular and molecular biology of cancer have helped in the identification of the mechanisms of cell malignant transformation, therefore allowing the use of new predictive factors and molecular treatments in cancer

Farmacogenómica: la medicina personalizada / Pharmacogenomic: the personalized medicine

El principal objetivo de la farmacogenómica (PGx) es definir un tratamiento farmacológico individualizado basado en el perfil genético de cada paciente, convirtiendo el paradigma clásico de tratamiento clínico centrado en la enfermedad en un nuevo enfoque, la medicina personalizada. Los polimorfismos genéticos pueden modificar la expresión y la función de las enzimas y las proteínas involucradas en la farmacocinética y la farmacodinámica de los fármacos. Así, la presencia de variantes alélicas permite predecir la respuesta farmacológica para garantizar la eficacia y la seguridad del tratamiento. Para la aplicación clínica de la PGx mediante la identificación de dichas variantes existen actualmente 2 planteamientos diferentes: el análisis de genes candidatos y los estudios de asociación genómica. La implementación clínica de la PGx mejora la eficacia, la seguridad y la relación costo-efectividad de los tratamientos; sin embargo, se ha ralentizado debido a una serie de barreras que se revisarán en este trabajo, así como sus posibles soluciones

The main objective of pharmacogenomics (PGx) is defining an individualized pharmacological treatment based on the genetic profile of each patient. Thus, the classical paradigm of clinical treatment focused on the disease is becoming a new approach, Personalized Medicine. The expression and function of enzymes and proteins involved in the drug pharmacokinetics and pharmacodynamics can be modified by genetic polymorphisms. Thereby, the presence of allelic variants allows predicting the pharmacological response guaranteeing the treatment efficacy and safety. Nowadays, two different approaches have been described for the clinical application of PGx by these variants identification: candidate gene analysis and genome wide association studies. Despite improving the effectiveness, safety and cost-effectiveness of treatments, the PGx clinical implementation has slowed down due to a series of barriers that will be reviewed in this work, as well as their possible solutions

The Future of Cardiovascular Computed Tomography: Advanced Analytics and Clinical Insights.

Cardiovascular computed tomography (CCT) has undergone rapid maturation over the last decade and is now of proven clinical utility in the diagnosis and management of coronary artery disease, in guiding structural heart disease intervention, and in the diagnosis and treatment of congenital heart disease. The next decade will undoubtedly witness further advances in hardware and advanced analytics that will potentially see an increasingly core role for CCT at the center of clinical cardiovascular practice. In coronary artery disease assessment this may be via improved hemodynamic adjudication, and shear stress analysis using computational flow dynamics, more accurate and robust plaque characterization with spectral or photon-counting CT, or advanced quantification of CT data via artificial intelligence, machine learning, and radiomics. In structural heart disease, CCT is already pivotal to procedural planning with adjudication of gradients before and following intervention, whereas in congenital heart disease CCT is already used to support clinical decision making from neonates to adults, often with minimal radiation dose. In both these areas the role of computational flow dynamics, advanced tissue printing, and image modelling has the potential to revolutionize the way these complex conditions are managed, and CCT is likely to become an increasingly critical enabler across the whole advancing field of cardiovascular medicine.

Prospective observational cohort study: Computational models for early prediction of ongoing pregnancy in fresh IVF/ICSI-ET protocols.

PURPOSE: This study sought to identify the significant factors related to ongoing pregnancy (OP) and to discover the most reliable model to distinguish OP from non-OP in early gestational age. METHODS: A total of 1650 cycles were enrolled in this study. Univariate Logistic Regression was used to identify the predictors included in multivariable analysis. The dataset was then randomly split into training set and test set with proportion of 70% and 30%. Forward stepwise multivariable logistic regression with 5-fold cross validation was used to build the final mathematic model. The performance of the model was determined by the arguments of test set. The area under receiver operating characteristic curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and misclassification rate (MR) were then calculated for model evaluation. RESULTS: Seven predictors were related to OP by univariate analysis. The serum hCG level on 14th day post-embryo-transfer (hCG14) and 21th day post-embryo-transfer (hCG21) were linear correlated. Therefore, different multivariate regression models were built using hCG14 or hCG21, respectively. After multivariate regression with 5-fold validation, the final indicators in model-1 were age_group, hCG21 and hCG21/hCG14, while age_group, hCG14, and calculated 48-hour-rising-ratio of hCG were the significant predictors in model-2. Model-2 showed better sensitivity and NPV, lower MR, and similar specificity and PPV. CONCLUSION: This study provided an effective mathematic model for early prediction of OP. The model could be of better clinical significance, especially for clinical counseling to manage patients' stress and anxiety, and for early warning of threatened miscarriage.

3D Printing and Heart Failure: The Present and the Future.

Advanced imaging modalities provide essential anatomic and spatial information in patients with complex heart disease. Two-dimensional imaging can be limited in the extent to which true 3-dimensional (3D) relationships are represented. The application of 3D printing technology has increased the creation of physical models that overcomes the limitations of a 2D screen. Many groups have reported the use of 3D printing for preprocedural planning in patients with different causes of heart failure. This paper reviews the innovative applications of this technique to provide patient-specific models to improve patient care.

El papel del laboratorio clínico en la medicina personalizada: situación actual y retos futuros / The role of the clinical laboratory in personalised medicine: Present situation and future challenges

La medicina personalizada, medicina de precisión o medicina individualizada ha sido definida como una manera de abordar el tratamiento y la prevención de las enfermedades en base a la variabilidad genética, ambiental y al estilo de vida de cada persona. Clasifica a los individuos en subpoblaciones que difieren en la susceptibilidad a desarrollar una enfermedad determinada o en la respuesta a un tratamiento específico, con el fin de aplicar el seguimiento y tratamiento más adecuado a cada paciente. La implementación de los procesos asociados a la Medicina Personalizada implica que los profesionales de laboratorio se enfrenten a una tecnología muy avanzada y poco conocida y a la dificultad de interpretación de los hallazgos, especialmente la valoración de su significación clínica. En este artículo se revisa la situación actual de la Medicina Personalizada, la función del laboratorio dentro de la misma y los retos que se deben afrontar

Personalised medicine, precision medicine, or individualised medicine has been defined as the way of preventing and treating diseases based on the genetic, environmental, and lifestyle variability for each individual. It classifies subjects into sub-populations that have different susceptibilities to develop a specific disease or to respond to a particular treatment. Its aim is to follow-up and treat each patient in the more suited to the patient. The establishment of the processes related to personalised medicine requires that specialists in Laboratory Medicine cope with cutting-edge, and little-known, technology with an interpretation that is highly complex from a clinical point of view. This review summarises the current situation of personalised medicine, the role of laboratory medicine in its implementation, and the challenges that need to be faced

The Various Applications of 3D Printing in Cardiovascular Diseases.

PURPOSE OF REVIEW: To highlight the various applications of 3D printing in cardiovascular disease and discuss its limitations and future direction. RECENT FINDINGS: Use of handheld 3D printed models of cardiovascular structures has emerged as a facile modality in procedural and surgical planning as well as education and communication. Three-dimensional (3D) printing is a novel imaging modality which involves creating patient-specific models of cardiovascular structures. As percutaneous and surgical therapies evolve, spatial recognition of complex cardiovascular anatomic relationships by cardiologists and cardiovascular surgeons is imperative. Handheld 3D printed models of cardiovascular structures provide a facile and intuitive road map for procedural and surgical planning, complementing conventional imaging modalities. Moreover, 3D printed models are efficacious educational and communication tools. This review highlights the various applications of 3D printing in cardiovascular diseases and discusses its limitations and future directions.

Medicina personalizada en atención primaria / Personalized medicine in primary care

No disponible

Systematic Review of Patient-Specific Surgical Simulation: Toward Advancing Medical Education.

OBJECTIVE: Simulation-based education has been shown to be an effective tool to teach foundational technical skills in various surgical specialties. However, most of the current simulations are limited to generic scenarios and do not allow continuation of the learning curve beyond basic technical skills to prepare for more advanced expertise, such as patient-specific surgical planning. The objective of this study was to evaluate the current medical literature with respect to the utilization and educational value of patient-specific simulations for surgical training. METHODS: We performed a systematic review of the literature using Pubmed, Embase, and Scopus focusing on themes of simulation, patient-specific, surgical procedure, and education. The study included randomized controlled trials, cohort studies, and case-control studies published between 2005 and 2016. Two independent reviewers (W.H.R. and N.D) conducted the study appraisal, data abstraction, and quality assessment of the studies. RESULTS: The search identified 13 studies that met the inclusion criteria; 7 studies employed computer simulations and 6 studies used 3-dimensional (3D) synthetic models. A number of surgical specialties evaluated patient-specific simulation, including neurosurgery, vascular surgery, orthopedic surgery, and interventional radiology. However, most studies were small in size and primarily aimed at feasibility assessments and early validation. CONCLUSIONS: Early evidence has shown feasibility and utility of patient-specific simulation for surgical education. With further development of this technology, simulation-based education may be able to support training of higher-level competencies outside the clinical settingto aid learners in their development of surgical skills.

Cardiac 3D Printing and its Future Directions.

Three-dimensional (3D) printing is at the crossroads of printer and materials engineering, noninvasive diagnostic imaging, computer-aided design, and structural heart intervention. Cardiovascular applications of this technology development include the use of patient-specific 3D models for medical teaching, exploration of valve and vessel function, surgical and catheter-based procedural planning, and early work in designing and refining the latest innovations in percutaneous structural devices. In this review, we discuss the methods and materials being used for 3D printing today. We discuss the basic principles of clinical image segmentation, including coregistration of multiple imaging datasets to create an anatomic model of interest. With applications in congenital heart disease, coronary artery disease, and surgical and catheter-based structural disease, 3D printing is a new tool that is challenging how we image, plan, and carry out cardiovascular interventions.

The Esophagiome: concept, status, and future perspectives.

The term "Esophagiome" is meant to imply a holistic, multiscale treatment of esophageal function from cellular and muscle physiology to the mechanical responses that transport and mix fluid contents. The development and application of multiscale mathematical models of esophageal function are central to the Esophagiome concept. These model elements underlie the development of a "virtual esophagus" modeling framework to characterize and analyze function and disease by quantitatively contrasting normal and pathophysiological function. Functional models incorporate anatomical details with sensory-motor properties and functional responses, especially related to biomechanical functions, such as bolus transport and gastrointestinal fluid mixing. This brief review provides insight into Esophagiome research. Future advanced models can provide predictive evaluations of the therapeutic consequences of surgical and endoscopic treatments and will aim to facilitate clinical diagnostics and treatment.

The Virtual Physiological Human: Ten Years After.

Biomedical research and clinical practice are struggling to cope with the growing complexity that the progress of health care involves. The most challenging diseases, those with the largest socioeconomic impact (cardiovascular conditions; musculoskeletal conditions; cancer; metabolic, immunity, and neurodegenerative conditions), are all characterized by a complex genotype-phenotype interaction and by a "systemic" nature that poses a challenge to the traditional reductionist approach. In 2005 a small group of researchers discussed how the vision of computational physiology promoted by the Physiome Project could be translated into clinical practice and formally proposed the term Virtual Physiological Human. Our knowledge about these diseases is fragmentary, as it is associated with molecular and cellular processes on the one hand and with tissue and organ phenotype changes (related to clinical symptoms of disease conditions) on the other. The problem could be solved if we could capture all these fragments of knowledge into predictive models and then compose them into hypermodels that help us tame the complexity that such systemic behavior involves. In 2005 this was simply not possible-the necessary methods and technologies were not available. Now, 10 years later, it seems the right time to reflect on the original vision, the results achieved so far, and what remains to be done.

Advanced technology in interventional cardiology: A roadmap for the future of precision coronary interventions.

Several specific new technologies [high-resolution CT coronary imaging with fractional flow reserve (CTCA-FFR), virtual reality (VR), vascular robotic systems (VRS), and three-dimensional printing] are poised to improve the treatment of patients with cardiovascular disease and at the same time the safety of the physicians who care for them. This article focuses on the potential clinical impact each of these modalities will have, as well as speculating on synergies that use of them together may achieve.

Left Atrial Appendage Closure Guided by 3D Printed Cardiac Reconstruction: Emerging Directions and Future Trends.

INTRODUCTION: Percutaneous left atrial appendage (LAA) occlusion has emerged as an alternative therapeutic approach to medical therapy for stroke prevention in patients with atrial fibrillation. 3D printing is a novel technology able to create a patient specific model of any given anatomical portion of the heart. RESULTS: Herein we report the first 2 cases of LAA occlusion procedure with 2 different systems, the Wave Crest device (Coherex Medical, Inc., USA) and the Amplatzer Amulet device (St. Jude Medical, St. Paul, MN, USA), in which a 3D printed LAA model (Care Tronik, Prato, Italy) was used in a rehearse phase. Both patients had history of paroxysmal AF and previous transient ischemic attack (TIA) occurred during oral anticoagulation with correct INR. In the first patient the occlusive device was positioned within the LAA after a rehearse occlusion using the 3D printed LAA plus a 27 mm Coherex Wavecrest device, demonstrating a good compression and sealing, particularly considering a proximal lobe of the appendage. In the second patient an attempt with the 27 mm Amulet device delivered within the 3D printed LAA, based on angiography and transesophageal echocardiographic (TEE), revealed insufficient covering of the proximal part of LAA vestibule; the device was released only after a second test with the 31 mm Amulet demonstrating a good sealing. CONCLUSION: These 2 cases demonstrated that 3D model could help in finding the correct position within LAA, sizing the device and guiding the choice of the closure device despite the measurements provided by angiography and TEE.