Immunoassays at a quartz-liquid interface: theory, instrumentation and preliminary application to the fluorescent immunoassay of human immunoglobulin G.

The theoretical basis and instrumental requirements of an optical detection technique for monitoring antibody-antigen reactions at a quartz-liquid interface are described. The antibody is covalently immobilized on the optical surface of a planar, fused-quartz waveguide and reacted with antigen solution. A light beam is internally reflected within the waveguide and penetrates into the solution only a fraction of the wavelength of the incident light. This is the evanescent wave which interacts optically with the growing number of antigen-antibody complexes but minimally with the bulk solution. A two-site immunofluorescent assay for human IgG measurement is described using fluorescein as the label. The assay detection limit is approximately 0.8 micrograms/ml and individual fluorescence measurements are completed within 10 min. It is expected that this evanescent wave immunoassay should have wide applicability in both routine and research fields.

Use of artificial intelligence in analytical systems for the clinical laboratory.

OBJECTIVE: To consider the role of software in system operation, control and automation, and attempts to define intelligence. METHODS AND RESULTS: Artificial intelligence (Al) is characterized by its ability to deal with incomplete and imprecise information and to accumulate knowledge. Expert systems, building on standard computing techniques, depend heavily on the domain experts and knowledge engineers that have programmed them to represent the real world. Neural networks are intended to emulate the pattern-recognition and parallel processing capabilities of the human brain and are taught rather than programmed. The future may lie in a combination of the recognition ability of the neural network and the rationalization capability of the expert system. In the second part of this paper, examples are given of applications of Al in stand-alone systems for knowledge engineering and medical diagnosis and in embedded systems for failure detection, image analysis, user interfacing, natural language processing, robotics and machine learning, as related to clinical laboratories. CONCLUSION: Al constitutes a collective form of intellectual property, and that there is a need for better documentation, evaluation and regulation of the systems already being used widely in clinical laboratories.

Dual-isotope separation technique for radioassay. Precision of calculated results affected by degree of separation.

The dual-isotope separation technique for radioassay is reviewed. An expression is derived describing the effect of counting errors on the precision of the calculated bound count rate in the dual-isotope technique. This mathematic model is tested in experiments using sodium iothalamate (I-125) as a marker in the Phadebas radiosorbent assay of cobalamin (Co-57). The coefficient of variation of results calculated on the basis of the dual-isotope technique is shown to be dependent on the amount of supernatant removed. The conclusion is that relatively large amounts of the supernatant must be removed before counting if the dual-isotope technique is to give acceptable results. The experimental model is proposed as a simple test of the suitability of a projected dual-isotope system.

Dual-isotope separation technique for radioassay of cobalamin.

Cobalamin is assayed by a dual-isotope separation method using sodium [125I]iothalamate as a marker. Two systems are used: one in which the incompletely-separated bound fraction is counted and compared with the single-isotope method in which the bound fraction is separated by washing (Phadebas radiosorbent assay); and one in which an aliquot of the free fraction is counted. In the dual-isotope method counting bound fractions, about 97% of the supernatant is removed by pouring from silicone fluid separators. The results for serum samples obtained using dual- and single-isotope methods were similar (between run coefficients of variation 5--7%). Experimental errors were smaller in the dual-isotope method. A factor in the kit standards, presumably the absence of proteins, was found to affect the separation technique, resulting in relatively large experimental errors for standards in the single-isotope method. Washing the solid phase in the single-isotope method apparently resulted in a loss of bound isotope. In the dual-isotope method counting free fractions reasonable precision was obtained (coefficient of variation of serum samples 6.6%) even though only about 56% of the supernatant (free fraction) was counted.

Dual isotope separation technique for radioassay. The effect of cross counting between two simultaneously counted isotopes.

A theoretical model, describing the effects of cross counting on the dual-isotope-corrected bound count rate and on the precision with which it is estimated, is presented. Experimental support for the validity of the model is given. The bound count rate is shown to be independent of the amount of marker isotope incorrectly counted as label isotope. The precision with which the bound count rate is estimated appears to be dependent on the size of this cross counting correction, although other factors such as pipetting errors are of more importance. Implications for the choice of assay system are discussed.

International Federation of Clinical Chemistry. Use of artificial intelligence in analytical systems for the clinical laboratory. IFCC Committee on Analytical Systems.

The incorporation of information-processing technology into analytical systems in the form of standard computing software has recently been advanced by the introduction of artificial intelligence (AI) both as expert systems and as neural networks. This paper considers the role of software in system operation, control and automation and attempts to define intelligence. AI is characterized by its ability to deal with incomplete and imprecise information and to accumulate knowledge. Expert systems, building on standard computing techniques, depend heavily on the domain experts and knowledge engineers that have programmed them to represent the real world. Neural networks are intended to emulate the pattern-recognition and parallel-processing capabilities of the human brain and are taught rather than programmed. The future may lie in a combination of the recognition ability of the neural network and the rationalization capability of the expert system. In the second part of this paper, examples are given of applications of AI in stand-alone systems for knowledge engineering and medical diagnosis and in embedded systems for failure detection, image analysis, user interfacing, natural language processing, robotics and machine learning, as related to clinical laboratories. It is concluded that AI constitutes a collective form of intellectual property and that there is a need for better documentation, evaluation and regulation of the systems already being used widely in clinical laboratories.

Opto-electronic immunosensors: a review of optical immunoassay at continuous surfaces.

Optical techniques for monitoring immunological reactions on continuous surfaces are reviewed. Initially Langmuir-Blodgett film techniques and ellipsometry are discussed, followed by internal reflection spectroscopy (IRS) systems. The latter includes attenuated total reflection (ATR) and total internal reflection fluorescence (TIRF). Finally, light scattering and surface plasmon resonance methods are presented. Overall, it was considered that the IRS systems and ellipsometric approaches offered the most promise for the design of a specific immunosensor device. Of these two, the ellipsometric methods are the most sensitive but also the most vulnerable to non-specific signal interference. Although lacking in extreme sensitivity, the IRS approaches reviewed were more specific in signal generation and were considered to have considerable potential for the future.

Optical detection of antibody-antigen reactions at a glass-liquid interface.

We describe an optical technique for detecting and monitoring antibody-antigen reactions at a solid-liquid interface. The antibody is covalently immobilized onto the surface of either a planar (microscope slide) or cylindrical (fibre optic) waveguide made of fused quartz. The reaction of immobilized antibody with antigen in solution is detected through use of the evanescent wave component of a light beam, which has a characteristic depth of penetration of a fraction of a wavelength into the aqueous phase, thus optically interacting primarily with substances bound (or located very close) to the interface and only minimally with the bulk solution. This resulting in-situ spatial separation of the antibody-bound from free antigen precludes a formal separation step and allows the reaction to be monitored kinetically. An immunoassay for methotrexate by absorption spectrometry achieved a detection limit of about 270 nmol/L; binding of methotrexate by immobilized antibody was monitored by the decrease in transmittance at 310 nm. A two-site immunofluorometric assay for human IgG could detect as little as 30 nmol/L; binding of fluorescein-labeled antibody was monitored by the increase in signal above 520 nm (lambda ex = 495 nm). With both immunoassays the signal-generating phase was monitored kinetically and was completed within 15 min.

Use of artificial intelligence in analytical systems for the clinical laboratory.

The incorporation of information-processing technology into analytical systems in the form of standard computing software has recently been advanced by the introduction of artificial intelligence (AI), both as expert systems and as neural networks.This paper considers the role of software in system operation, control and automation, and attempts to define intelligence. AI is characterized by its ability to deal with incomplete and imprecise information and to accumulate knowledge. Expert systems, building on standard computing techniques, depend heavily on the domain experts and knowledge engineers that have programmed them to represent the real world. Neural networks are intended to emulate the pattern-recognition and parallel processing capabilities of the human brain and are taught rather than programmed. The future may lie in a combination of the recognition ability of the neural network and the rationalization capability of the expert system.In the second part of the paper, examples are given of applications of AI in stand-alone systems for knowledge engineering and medical diagnosis and in embedded systems for failure detection, image analysis, user interfacing, natural language processing, robotics and machine learning, as related to clinical laboratories.It is concluded that AI constitutes a collective form of intellectual propery, and that there is a need for better documentation, evaluation and regulation of the systems already being used in clinical laboratories.