Diagnostics 4.0: the medical laboratory in digital health

Machine learning-based clinical decision support using laboratory data

Artificial intelligence (AI) and machine learning (ML) are becoming vital in laboratory medicine and the broader context of healthcare. In this review article, we summarized the development of ML models and how they contribute to clinical laboratory workflow and improve patient outcomes. The process of ML model development involves data collection, data cleansing, feature engineering, model development, and optimization. These models, once finalized, are subjected to thorough performance assessments and validations. Recently, due to the complexity inherent in model development, automated ML tools were also introduced to streamline the process, enabling non-experts to create models. Clinical Decision Support Systems (CDSS) use ML techniques on large datasets to aid healthcare professionals in test result interpretation. They are revolutionizing laboratory medicine, enabling labs to work more efficiently with less human supervision across pre-analytical, analytical, and post-analytical phases. Despite contributions of the ML tools at all analytical phases, their integration presents challenges like potential model uncertainties, black-box algorithms, and deskilling of professionals. Additionally, acquiring diverse datasets is hard, and models' complexity can limit clinical use. In conclusion, ML-based CDSS in healthcare can greatly enhance clinical decision-making. However, successful adoption demands collaboration among professionals and stakeholders, utilizing hybrid intelligence, external validation, and performance assessments.

Specific test panels for patients with heart failure: implementation and use in the Spanish National Health System

Objectives

The use of specific test panels (STP) for heart failure (HF) could help improve the management of this condition. The purpose of this study is to gain an insight into the level of implementation of STPs in the management of HF in Spain and gather the opinions of experts, with a special focus on parameters related to iron metabolism.

Methods

The opinions of experts in HF were gathered in three stages STAGE 1 as follows: level of implementation of STPs (n=40). STAGE 2: advantages and disadvantages of STPs (n=12). STAGE 3: level of agreement with the composition of three specific STPs for HF: initial evaluation panel, monitoring panel, and de novo panel (n=16).

Results

In total, 62.5% of hospitals used STPs for the clinical management of HF, with no association found between the use of STPs and the level of health care (p=0.132) and location of the center (p=0.486) or the availability of a Heart Failure Unit in the center (p=0.737). According to experts, the use of STPs in clinical practice has more advantages than disadvantages (8 vs. 3), with a notable positive impact on diagnostics. Experts gave three motivations and found three limitations to the implementation of STPs. The composition of the three specific STPs for HF was viewed positively by experts.

Conclusions

Although the experts interviewed advocate the use of diagnostic and monitoring STPs for HF, efforts are still necessary to achieve the standardization and homogenization of test panels for HF in Spanish hospitals.

Has the Flood Entered the Basement? A Systematic Literature Review about Machine Learning in Laboratory Medicine

This article presents a systematic literature review that expands and updates a previous review on the application of machine learning to laboratory medicine. We used Scopus and PubMed to collect, select and analyse the papers published from 2017 to the present in order to highlight the main studies that have applied machine learning techniques to haematochemical parameters and to review their diagnostic and prognostic performance. In doing so, we aim to address the question we asked three years ago about the potential of these techniques in laboratory medicine and the need to leverage a tool that was still under-utilised at that time.

Development and evaluation of point-of-care testing recertification with e-learning

One of the main requirements in point-of-care testing (POCT) is efficient operator training to avoid diagnostic errors. Considering a variety of users and time-independent learning, e-learning is preferred. However, in our experience, e-learning is not always accepted by employees. After using a commercial e-learning program with little success, we developed a specific e-learning offer to achieve a better acceptance of online-based training. Herein our goal was to identify the most relevant aspects for better acceptance. The new e-learning modules were implemented with the learning management system ILIAS and dealt with typical sources of error. The implementation was accompanied by an anonymous online questionnaire within the POCT operators examining differences between the acceptance of the commercial e-learning and the hospital-specific. The results show higher acceptance for clinic-specific e-learning whereby online training of the POCT operators could successfully established. Most relevant aspects are the relevance of contents for the personal work and the working processes within the clinic as well as processing time. Thereby, the recertification of the POCT operators based on the successful completion of the learning modules was fully integrated in the POCT process. In respect to the need for regular recertification of POCT operators, our study shows that the acceptance of e-learning could be improved by adapting e-learning modules to the specific workflows in the hospital.