POCT especially important for patients suffering from critical conditions, as they can monitor their health constantly without the need to go to hospital, or for trained personnel. Artifi cial intelligence and machine learning Artifi cial intelligence (AI), machine learning (ML) and neural networks are now starting to be integrated into POCT. These AI modules have demonstrated their value in diagnostics studies. However, the accuracy of test apps varies greatly and relying on these apps is cautioned against. The future is bright. Since samples can be digitised directly at the point-of-care (POC), advanced digital diagnostic techniques, such as sample analysis using medical AI algorithms, can be deployed outside high-end laboratories. “Hence POC digital microscopy, supported by automated digital image analysis and AI, might be deployed for routine microscopy diagnostics on samples harvested during cancer surgery, or on parasitology samples with an emphasis on potential areas of application in low-resource settings,” added Oscar Holmström from the Faculty of Medicine at the University of Helsinki. r e h p a r g o t p y r C / k c o t s r e t t u h S © POCT and Covid-19 Most promising SARS-CoV-2 POCT methods: immunoas- says for antibody and antigen detection, RT-PCR as the gold standard, isothermal amplifi cation and CRISPR / Cas9 genome editing technology as an emerging technique. Due to growing confi rmed COVID-19 cases globally, rapid and reliable POCT is needed urgently for early detection. A reliable POCT device could reduce transportation, risk of spreading infection, strain on healthcare, and cost of care. Despite outbreaks caused by infectious viral diseases such as MERS, SARS, and Ebola, existing POCT platforms were not suffi ciently ready to address the COVID-19 viral threat. However, in 2020 great efforts were made in POCT to improve COVID-19 detection. LABBook 2021 89 Technology advances change POCT applications Miniaturisation in chip technology, microfl uidics and new biosensors have heralded the design of new systems for POCT. Lab-on-a-chip technology is one of the main drivers of POCT, especially in infectious disease diagnosis. This technology means various bioassays, such as microbiological culture, polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay (ELISA), can now be used at the point of care. Similarly, emerging microfl uidic technologies include a set of miniaturised components allowing chemical or biological samples to be analysed at the microscopic level. Microfl uidic-based POCT devices are widely used in molecular biology as well as in chemi- cal and biochemical analysis. They enable detection and fl uid reg- ulation in a single unit. “Greater sensitivity and specifi city when detecting target analytes in small volumes overcomes several challenges encountered when using traditional POCT techniques,” Saha said. POCT currently centres around two technologies: lateral fl ow assay (LFA) and nucleic acid amplifi cation. The fi rst is used in pregnancy testing; testing for HIV, herpes simplex virus, hepatitis, infectious diseases (Ebola, dengue, malaria, Zika virus) and respiratory infections; and for diagnosis and prognosis in conditions such as cancer, by identifying specifi c biomarkers. The second, much more sensitive and specifi c, is based on polymerase chain reactions (PCR) on a chip and isothermal amplifi cation. Nucleic acid amplifi cation can be used to detect a whole array of infectious diseases, such as Mycoplasma pneumonia, Bordetella pertussis, Legionella pneumonia, Infl uenza A virus, SARS, Legio- nella, Aspergillus, West Nile Virus and, now, SARS-CoV-2. A seventh format joins POCT POC device manufacturers are continually looking for ways to design products that deliver greater user comfort in a cost-effective manner. POCT relies on six main formats: bench-top, monitoring, transportable, portable, handheld and disposable. In recent years, the latest advances have seen the launch of a seventh: smart devices with smartphones and wearable devices. Mobile POCT uses sensors to detect signals from samples in vitro, whereas wearable POCT detects signals directly on the body. Both systems then send quantifi ed results to the clinic via wireless communication. Various body fl uids such as tears, urine, blood, sweat and saliva can be used to analyse metabolites, hormones, proteins, viruses and bacteria. “Smartphones act as minicomputers for sensitive and specifi c data quantifi cation with built-in sensors, high resolution cameras, rapid wireless connectivity and the ability to use various software and apps. This means they can function as standalone sensors and detectors in mobile POCT,” Saha pointed out. Similarly, wearable POCT devices can be physical sensors used to acquire samples from the skin, eye or mouth with minimal invasion. They come in various forms, such as tattoos, patches, bands, watches, spectacles and contact lenses, and can be inte- grated with smartphones for data capture. This type of testing is