Expert systems in histopathology. II. Knowledge representation and rule-based systems

Two aspects of expert systems for use in diagnostic histopathology and cytopathology are examined: knowledge representation and the structure and operation of rule-based systems. Knowledge may be represented, e.g., in semantic networks, frames, multiple contexts and model-based structures; the choice of structure should be matched to the type of information to create an efficient and logically adequate expert system. In a rule-based system, knowledge is represented as "rules," often in the form of "IF (condition)-THEN (conclusion)" rules. The anatomy of such rules and their operation is explored via the use of examples. Uncertainty in rules is briefly addressed, and their processing by the symbolic reasoning of the "inference engine" of the expert system is described, including both "forward-chaining" ("data-driven") operations and "backward-chaining" ("goal-driven") operations.     

The parametric pump mechanism in separation of components in heterogeneous systems. I. Macroscopic distributed systems.

A dynamic method is proposed for the separation of the electrolyte components using a parametric pump with an ion exchange column. It was studied experimentally and described mathematically. The parametric separation of mixtures is based on interactions of two oscillating fields with a heterogeneous system containing two phases, a liquid and a solid one, the components of the mixture being able to redistribute between the phases. The field of mechanical force is responsible for cyclic relative displacement of the phases, and synchronously changing temperature causes redistribution of the components between them. This results in sodium and potassium fluxes opposite in direction which in turn leads to accumulation of sodium and potassium in opposite end cells.     

Knowledge systems.

The primate brain appears to store knowledge in widely distributed neural systems. Recent evidence suggests that these systems are specialized with respect to their capacity for supporting acquisition and retrieval of different domains and levels of knowledge.     

Bimolecular systems: III. Catalytic systems

Amplification effect in the catalytic bimolecular systems is a consequence of the kinetic characteristic of the catalyst. Two types of the coefficient of amplification are defined. The applicability of these definitions is given by the type of the bimolecular system. In a simple example it is shown that the concept of amplification is meaningful in these systems. Furthermore, two rules, analogous to those for a coupling of amplifiers, are derived for the two basic modes of coupling of catalytic systems. Thus, in biological systems the catalytic reactions may be regarded as biologically effective amplifiers.     

Bimolecular systems: I. Linear systems of complexes

A vast number of biologically important processes are based upon bimolecular systems. In these systems intermediate complexes are formed. Bimolecular systems in which no complex-complex interactions occur are called linear systems of complexes. A definition and some characteristic properties of these systems are given here. There may exist a contradiction of Onsager's principle of detailed balancing in these systems; however, no principal differences are found between the steady state behavior of an open system and that of a closed system. It is shown that the steady state behavior of a linear system of complexes of arbitrary complexity has some similarities with the steady state behavior of a simple bimolecular system, e.g., Michaelis-Menten enzymatic reaction. Multiplicity of action of the substances participating in biomolecular processes may produce some qualitative differences in the steady state behavior of the system.     

Multiple excitations in photosynthetic systems.

The yield of fluorescence in Chlorella from a 7 ns pulse of light is found to decrease gradually as a function of the number of hits in the photosynthetic units. The fivefold decrease in yield is spread over some three orders of magnitude of pulse energy and strongly suggests another random process in addition to that of photon absorption. Evidence supports the view that this random process is not in the time but in the spatial domain. The model used to fit the data is that of a unit with multiple traps for the singlet excitation. An excitation is captured by an open trap or destroyed by a filled trap with equal probability. These studies give evidence for the connectivity of the photosynthetic energy transfer apparatus on the short time scale. The short fluorescence lifetimes following picosecond pulse excitation of photosynthetic systems reported by several laboratories may be explained by the effect of multiple excitations.     

Bioconversions in aqueous two-phase systems.

Bioconversions involving enzymes and/or microbial cells in aqueous two-phase systems are reviewed. The partitioning of biocatalysts, substrates, and products is discussed in relation to their size. The efficiency of retaining biocatalysts in aqueous two-phase systems is summarized in relation to other methods of recirculating. The influence of phase components on the activity and the stability of enzymatic biocatalysts is exemplified with penicillin acylase and the cellulolytic enzyme system, and the effect of phase components on biocatalytic living cells is exemplified with the production of alpha-amylase with Bacillus sp. Process design costs in bioconversions in aqueous two-phase systems are briefly summarized.     

Synaptic plasticity in cortical systems.

Recent studies indicate that synapse addition and/or loss is associated with different types of learning. Other factors influencing synaptogenesis and synapse loss include neurotrophins, hormones, and the induction of long-term potentiation. An emerging view of synaptic plasticity suggests that local neurotrophin action and synaptically associated protein synthesis may promote synaptic remodelling and changes in receptor expression or activation.     

Bimolecular systems: II. Nonlinear systems of complexes

This paper deals with bimolecular systems in which also complex-complex interactions occur. Because of the complexity of the problem, an approximation in a form of coupled linear systems of complexes (Bull. Math. Biophysics,29, 1–16, 1967) is considered. Two types of couplings, serial and parallel, are studied. In the serial coupling the nonlinear system of complexes has the same behavior as its subsystems. An entity, initial sensitivity, has interesting properties: in serial coupling it is at most equal to the product and in parallel coupling, at most equal to the sum of partial initial sensitivities.