Validation of a novel Artificial Pharmacological Intelligence (API) system for the management of patients with polypharmacy

OBJECTIVE

Medication management of patients with polypharmacy is highly complex. We aimed to validate a novel Artificial Pharmacological Intelligence (API) algorithm to optimize the medication review process in a comprehensive, personalized, and scalable way.

MATERIALS AND METHODS

The study was conducted on anonymized retrospective electronic health records (EHR) of 49 patients. Each patient's file was reviewed by the API system, a clinical pharmacist, and a judging committee. Validation was assessed by comparing the overall agreement of the judging committee (as the gold standard, blinded to the identity of the analyzer) to both the API system and clinical pharmacists' conclusions. Five medication-related problem (MRP) categories were assessed: duplication of therapy, age-related issues, incorrect dose, current side effects and future side effects' risk. For each category the overall validity parameters, agreement, positive predictive value (PPV), negative predictive value (NPV), sensitivity and specificity were analyzed.

RESULTS

The agreement between the API system and the judging committee was 93.5 % (95 % CI 92.7-94.4), while the agreement between the clinical pharmacists and the judging committee was 73.9 % (95 % CI 72.5-75.3). The PPV was 92.2 % (90.9-93.5) and NPV was 94.2 % (93.1-95.2) for the API system and 76.3 % (69.8-82.8) and 73.5 % (72.3-74.8) respectively for the clinical pharmacists.

DISCUSSION

AI systems can equip clinicians with sophisticated tools and scale manual processes such as comprehensive medication reviews, thus reducing MRPs and drug-related hospitalizations related to multidrug treatments. The API system validated in this study provided comprehensive, multidrug, multilayered analysis intended to bridge the innate complexity of personalized polypharmacy treatment.

CONCLUSIONS

The API system was validated as a tool for providing actionable clinical insights non-inferior to a manual clinical review of a clinical pharmacist. The API system showed promising results in reducing MRPs.

On-cloud decision-support system for non-small cell lung cancer histology characterization from thorax computed tomography scans

Non-Small Cell Lung Cancer (NSCLC) accounts for about 85% of all lung cancers. Developing non-invasive techniques for NSCLC histology characterization may not only help clinicians to make targeted therapeutic treatments but also prevent subjects from undergoing lung biopsy, which is challenging and could lead to clinical implications. The motivation behind the study presented here is to develop an advanced on-cloud decision-support system, named LUCY, for non-small cell LUng Cancer histologY characterization directly from thorax Computed Tomography (CT) scans. This aim was pursued by selecting thorax CT scans of 182 LUng ADenocarcinoma (LUAD) and 186 LUng Squamous Cell carcinoma (LUSC) subjects from four openly accessible data collections (NSCLC-Radiomics, NSCLC-Radiogenomics, NSCLC-Radiomics-Genomics and TCGA-LUAD), in addition to the implementation and comparison of two end-to-end neural networks (the core layer of whom is a convolutional long short-term memory layer), the performance evaluation on test dataset (NSCLC-Radiomics-Genomics) from a subject-level perspective in relation to NSCLC histological subtype location and grade, and the dynamic visual interpretation of the achieved results by producing and analyzing one heatmap video for each scan. LUCY reached test Area Under the receiver operating characteristic Curve (AUC) values above 77% in all NSCLC histological subtype location and grade groups, and a best AUC value of 97% on the entire dataset reserved for testing, proving high generalizability to heterogeneous data and robustness. Thus, LUCY is a clinically-useful decision-support system able to timely, non-invasively and reliably provide visually-understandable predictions on LUAD and LUSC subjects in relation to clinically-relevant information.

A cluster randomized controlled trial of an electronic decision-support system to enhance antenatal care services in pregnancy at primary healthcare level in Telangana, India: trial protocol

BACKGROUND

India contributes 15% of the total global maternal mortality burden. An increasing proportion of these deaths are due to Pregnancy Induced Hypertension (PIH), Gestational Diabetes Mellitus (GDM), and anaemia. This study aims to evaluate the effectiveness of a tablet-based electronic decision-support system (EDSS) to enhance routine antenatal care (ANC) and improve the screening and management of PIH, GDM, and anaemia in pregnancy in primary healthcare facilities of Telangana, India. The EDSS will work at two levels of primary health facilities and is customized for three cadres of healthcare providers - Auxiliary Nurse Midwifes (ANMs), staff nurses, and physicians (Medical Officers).

METHODS

This will be a cluster randomized controlled trial involving 66 clusters with a total of 1320 women in both the intervention and control arms. Each cluster will include three health facilities-one Primary Health Centre (PHC) and two linked sub-centers (SC). In the facilities under the intervention arm, ANMs, staff nurses, and Medical Officers will use the EDSS while providing ANC for all pregnant women. Facilities in the control arm will continue to provide ANC services using the existing standard of care in Telangana. The primary outcome is ANC quality, measured as provision of a composite of four selected ANC components (measurement of blood pressure, blood glucose, hemoglobin levels, and conducting a urinary dipstick test) by the healthcare providers per visit, observed over two visits. Trained field research staff will collect outcome data via an observation checklist.

DISCUSSION

To our knowledge, this is the first trial in India to evaluate an EDSS, targeted to enhance the quality of ANC and improve the screening and management of PIH, GDM, and anaemia, for multiple levels of health facilities and several cadres of healthcare providers. If effective, insights from the trial on the feasibility and cost of implementing the EDSS can inform potential national scale-up. Lessons learned from this trial will also inform recommendations for designing and upscaling similar mHealth interventions in other low and middle-income countries.

TRIAL REGISTRATION

CLINICALTRIALS

gov, NCT03700034, registered 9 Oct 2018, https://www.

CLINICALTRIALS

gov/ct2/show/NCT03700034 CTRI, CTRI/2019/01/016857, registered on 3 Mar 2019, http://www.ctri.nic.in/Clinicaltrials/pdf_generate.php?trialid=28627&EncHid=&modid=&compid=%27,%2728627det%27.

A clinical decision-support system for dengue based on fuzzy cognitive maps

Dengue is a viral infection widely distributed in tropical and subtropical regions of the world. Dengue is characterized by high fatality rates when the diagnosis is not made promptly and effectively. To aid in the diagnosis of dengue, we propose a clinical decision-support system that classifies the clinical picture based on its severity, and using causal relationships evaluates the behavior of the clinical and laboratory variables that describe the signs and symptoms related to dengue. The system is based on a fuzzy cognitive map that is defined by the signs, symptoms and laboratory tests used in the conventional diagnosis of dengue. The evaluation of the model was performed on datasets of patients diagnosed with dengue to compare the model with other approaches. The developed model showed a good classification performance with 89.4% accuracy and could evaluate the behaviour of clinical and laboratory variables related to dengue severity (it is an explainable method). This model serves as a diagnostic aid for dengue that can be used by medical professionals in clinical settings.

Decision-support systems for managing polypharmacy in the elderly: A scoping review

Polypharmacy, the consuming of more than five drugs, is a public health problem. It can lead to many interactions and adverse drug reactions and is very expensive. Therapeutic guidelines for managing polypharmacy in the elderly have been issued, but are highly complex, limiting their use. Decision-support systems have therefore been developed to automate the execution of these guidelines, or to provide information about drugs adapted to the context of polypharmacy. These systems differ widely in terms of their technical design, knowledge sources and evaluation methods. We present here a scoping review of electronic systems for supporting the management, by healthcare providers, of polypharmacy in elderly patients. Most existing reviews have focused mainly on evaluation results, whereas the present review also describes the technical design of these systems and the methodologies for developing and evaluating them. A systematic bibliographic search identified 19 systems differing considerably in terms of their technical design (rule-based systems, documentary approach, mixed); outputs (textual report, alerts and/or visual approaches); and evaluations (impact on clinical practices, impact on patient outcomes, efficiency and/or user satisfaction). The evaluations performed are minimal (among all the systems identified, only one system has been evaluated according to all the criteria mentioned above) and no machine learning systems and/or conflict management systems were retrieved. This review highlights the need to develop new methodologies, combining various approaches for decision support system in polypharmacy.

Participatory implementation of an antibiotic stewardship programme supported by an innovative surveillance and clinical decision-support system

BACKGROUND

Antibiotic resistance will cause about 10 million deaths per year by 2050. Fighting antimicrobial resistance is a health priority. Interventions aimed to reduce antimicrobial resistance, such as antibiotic stewardship programmes (ASPs), must be implemented. To be effective, those interventions, and the implementation process, should be matched with social-cultural context. The complexity of ASPs can no longer be developed without considering both organizational and information systems.

AIM

To support ASPs through the co-design and implementation, in collaboration with healthcare workers, of a surveillance and clinical decision-support system to monitor antibiotic resistance and improve antibiotic prescription.

METHODS

The surveillance and clinical decision-support system was designed and implemented in three Portuguese hospitals, using a participatory approach between researchers and healthcare workers following the Design Science Research Methodology.

FINDINGS

Based on healthcare workers' requirements, we developed HAITooL, a real-time surveillance and clinical decision-support system that integrates visualizations of patient, microbiology, and pharmacy data, facilitating clinical decision. HAITooL monitors antibiotic usage and rates of antibiotic-resistant bacteria, allowing early identification of outbreaks. It is a clinical decision-support tool that integrates evidence-based algorithms to support proper antibiotic prescription. HAITooL was considered valuable to support monitoring of antibiotic resistant infections and an important tool for ASP sustainability.

CONCLUSION

ASP implementation can be leveraged through a surveillance and clinical decision-support system such as HAITooL that allows antibiotic resistance monitoring and supports antibiotic prescription, once it has been adapted to the context and specific needs of healthcare workers and hospitals.

Applying the Zone of Proximal Development when Evaluating Clinical Decision Support Systems: A Case Study

The goal to facilitate a continuing medical education can be incorporated in the design of a clinical decision-support system. Developing a method for evaluating knowledge and skill development as part of evaluating the system is the aim for the research presented in this paper. The activity supported by the system was analyzed using Activity theory and structured into a protocol. Four clinicians were studied using the system for the first time, and their activity were assessed using the concept of Zone of Proximal Development. Initial results show how the system was used for clinician with different level of skills, and provide implications for further development of the methodology and the system.