Applications of machine learning in the chemical pathology laboratory

Machine learning (ML) is an area of artificial intelligence that provides computer programmes with the capacity to autodidact and learn new skills from experience, without continued human programming. ML algorithms can analyse large data sets quickly and accurately, by supervised and unsupervised learning techniques, to provide classification and prediction value outputs. The application of ML to chemical pathology can potentially enhance efficiency at all phases of the laboratory’s total testing process. Our review will broadly discuss the theoretical foundation of ML in laboratory medicine. Furthermore, we will explore the current applications of ML to diverse chemical pathology laboratory processes, for example, clinical decision support, error detection in the preanalytical phase, and ML applications in gel-based image analysis and biomarker discovery. ML currently demonstrates exploratory applications in chemical pathology with promising advancements, which have the potential to improve all phases of the chemical pathology total testing pathway.

Error Evaluation in the Laboratory Testing Process and Laboratory Information Systems

Introduction: The laboratory testing process consists of five analysis phases featuring the total testing process framework. Activities in laboratory process, including those of testing, are error-prone and affect the use of laboratory information systems. This study seeks to identify error factors related to system use and the first and last phases of the laboratory testing process using a proposed framework known as total testing process-laboratory information systems. Materials and Methods: We conducted a qualitative case study in two private hospitals and a medical laboratory. We collected data using interviews, observations, and document analysis methods involving physicians, nurses, an information technology officer, and the laboratory staff. We employed the proposed framework and Lean problem-solving tools namely Value Stream Mapping and A3 for data analysis. Results: Errors in laboratory information systems and the laboratory testing process were attributed to failure to fulfill user requirements, poor cooperation between the information technology unit and laboratory, the inconsistency of software design in system integration, errors during inter-system data transmission, and lack of motivation in system use. The error factors are related to system development elements, namely, latent failures that considerably affected the information quality and system use. Errors in system development were also attributed to poor service quality. Conclusion: Complex laboratory testing process and laboratory information systems require rigorous evaluation in minimizing errors and ensuring patient safety. The proposed framework and Lean approach are applicable for evaluating the laboratory testing process and laboratory information systems in a rigorous, comprehensive, and structured manner.

Development and Application of Computerized Risk Registry and Management Tool Based on FMEA and FRACAS for Total Testing Process

Background and Objectives: Risk management is considered an integral part of laboratory medicine to assure laboratory quality and patient safety. However, the concept of risk management is philosophical, so actually performing risk management in a clinical laboratory can be challenging. Therefore, we would like to develop a sustainable, practical system for continuous total laboratory risk management. Materials and Methods: This study was composed of two phases: the development phase in 2019 and the application phase in 2020. A concept flow diagram for the computerized risk registry and management tool (RRMT) was designed using the failure mode and effects analysis (FMEA) and the failure reporting, analysis, and corrective action system (FRACAS) methods. The failure stage was divided into six according to the testing sequence. We applied laboratory errors to this system over one year in 2020. The risk priority number (RPN) score was calculated by multiplying the severity of the failure mode, frequency (or probability) of occurrence, and detection difficulty. Results: 103 cases were reported to RRMT during one year. Among them, 32 cases (31.1%) were summarized using the FMEA method, and the remaining 71 cases (68.9%) were evaluated using the FRACAS method. There was no failure in the patient registration phase. Chemistry units accounted for the highest proportion of failure with 18 cases (17.5%), while urine test units accounted for the lowest portion of failure with two cases (1.9%). Conclusion: We developed and applied a practical computerized risk-management tool based on FMEA and FRACAS methods for the entire testing process. RRMT was useful to detect, evaluate, and report failures. This system might be a great example of a risk management system optimized for clinical laboratories.

Automation of extra-analytical phase for clinical laboratory

Extra-analytical automation is of critical importance in patient safety with respect to accurate, fast test result reporting. Through the previous decades, significant improvements in laboratory errors have been achieved by technological facilities, which have become a substantial part of the reduction of preventable diagnostic errors. In clinical laboratory practice, the total testing process (TTP) is under the effect of error sources: preanalytical, analytical, and post-analytical variables. Since many extra-analytical processes within and outside the clinical laboratory may be automated, management of the extra-analytical phase can prevent errors, resulting in the total quality of laboratory diagnostics and customer satisfaction. The automation technologies have added a serious impact on the proficiency of clinical laboratories. To improve standardization, organization, efficiency, and quality of TTP, many manual tasks have now been partially or entirely automated by labor-saving instrumentations. The implementation of extra-analytical automation in the laboratory processes has recently made them standardized and manageable. Depending on the workload and workflow of the clinical laboratory, it is of critical importance to implement adequate systems, providing standardization of the TTP and resulting in high-quality test results.

How to meet ISO15189:2012 pre-analytical requirements in clinical laboratories? A consensus document by the EFLM WG-PRE

The International Organization for Standardization (ISO) 15189:2012 standard aims to improve quality in medical laboratories through standardization of all key elements in the total testing process, including the pre-analytical phase. It is hence essential that accreditation bodies, assessing laboratories against ISO15189:2012, pay sufficient attention to auditing pre-analytical activities. However, there are significant differences in how technical auditors interpret the pre-analytical requirements described in ISO15189:2012. In this consensus document, the European Federation of Clinical Chemistry and Laboratory Medicine (EFLM) Working Group for Pre-analytical Phase (WG-PRE) sets out to review pre-analytical requirements contained in ISO15189:2012 and provide guidance for laboratories on how to meet these requirements. The target audience for this consensus document is laboratory professionals who wish to improve the quality of the pre-analytical phase in their laboratory. For each of the ISO requirements described in ISO15189:2012, members of EFLM WG-PRE agreed by consensus on minimal recommendations and best-in-class solutions. The minimal consensus recommendation was defined as the minimal specification which laboratories should implement in their quality management system to adequately address the pre-analytical requirement described in ISO15189:2012. The best-in-class solution describes the current state-of-the-art in fulfilling a particular pre-analytical requirement in ISO15189:2012. We fully acknowledge that not every laboratory has the means to implement these best-in-class solutions, but we hope to challenge laboratories in critically evaluating and improving their current procedures by providing this expanded guidance.

Policies and practices in the field of laboratory hematology in Croatia – a current overview and call for improvement

Objectives In 2019 The Croatian Working Group for Laboratory Hematology, on behalf of the Croatian Society of Medical Biochemistry and Laboratory Medicine, wanted to explore the background in field of laboratory hematology routine practice among Croatian laboratories in order to develop future strategies for producing national recommendations, if needed. Methods During April and May 2019, a comprehensive survey covering all main parts of the total testing process within the field of laboratory hematology among Croatian medical laboratories was conducted. The survey comprised 49 inquiries. Data was collected using Survey Monkey (Palo Alto, CA, USA). All collected data was anonymized. Results The response rate was 72%. There is still a substantial number of laboratories that have only three-part differential hematology analyzers (9%). Furthermore, a very high number of laboratories did not perform analyzer verification prior to implementation into routine work (31%). Out of those who have verified their analyzers, a diversity of guidelines and recommendations were used. Nearly 10% of the laboratories do not have a defined policy regarding specimen rejection. The majority of the participants perform internal quality control daily (83%), however, only 51% of respondents evaluate the agreement between different hematology analyzers on daily basis. Although more than 90% of Croatian laboratories have a defined policy regarding specimen rejection, only 61% of respondents continuously monitor quality indicators in routine practice. Conclusions The survey revealed substantial differences in all aspects of laboratory hematology practices among Croatian medical laboratories, indicating the need for universal recommendations at the national level.

Patient Safety in Laboratory Medicine

Laboratory medicine in the healthcare system has recently been recognized as a fundamental service in the clinical decision-making process. Therefore, the notion of patient safety in laboratory medicine must be recognized as the assurance that harm to patients will be avoided, safe care outcomes will be enhanced through error prevention, and the total testing process (TTP) will be continuously improved.Although the goal for patient safety is zero errors, and although laboratory professionals have made numerous efforts to reduce errors in the last few decades, current research into laboratory-related diagnostic errors highlights that: (a) errors occur at every step of the TTP, mainly affecting phases at clinical interfaces; (b) despite the improvement strategies adopted, analytical quality remains a challenge; (c) errors are linked not only to clinical chemistry tests, but also to new, increasingly complex diagnostic testing.Medical laboratories must therefore implement effective quality assurance tools to identify and prevent errors in order to guarantee the reliability of laboratory information. Accreditation in compliance with the International Standard ISO 15189 represents the first step, establishing processes with excellence requirements and greater expectations of staff competency. Another important step in preventing errors and ensuring patient safety is the development of specific educational and training programs addressed to all professionals involved in the process, in which both technical and administrative skills are integrated. A wide variety of information is provided by a robust quality management system and consensus-approved Quality Indicators (QI) that identify undesirable events, evaluate the risk to the patient, and call for corrective and preventive actions. However, the effectiveness of the system depends on the careful analysis of data collected and on staff awareness of the importance of laboratory medicine to the healthcare process. The main task of the new generation of laboratory professionals should be to gain experience in “clinical laboratory stewardship.” In order to safeguard patients, laboratory professionals must assist clinicians in selecting the right test for the right patient at the right time and facilitate the interpretation of laboratory information.