PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 1889199

A model for diagnosing and explaining multiple disorders.

P W Jamieson
View original source ↗Export metadata ↓

Abstract

The ability to diagnose multiple interacting disorders and explain them in a coherent causal framework has only partially been achieved in medical expert systems. This paper proposes a causal model for diagnosing and explaining multiple disorders whose key elements are: physician-directed hypotheses generation, object-oriented knowledge representation, and novel explanation heuristics. The heuristics modify and link the explanations to make the physician aware of diagnostic complexities. A computer program incorporating the model currently is in use for diagnosing peripheral nerve and muscle disorders. The program successfully diagnoses and explains interactions between diseases in terms of underlying pathophysiologic concepts. The model offers a new architecture for medical domains where reasoning from first principles is difficult but explanation of disease interactions is crucial for the system's operation.

Explore related subjects

Acid-Base ImbalanceAdultArtificial IntelligenceDiagnosis, Computer-AssistedHumansMaleModels, BiologicalSoftwareWater-Electrolyte ImbalanceCase ReportsJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

P W Jamieson. 1991. A model for diagnosing and explaining multiple disorders.. https://doi.org/10.1016/0010-4809(91)90031-q

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1992

Blood flow velocity waveforms of the middle cerebral artery and abnormal neurological evaluations in live-born fetuses with absent or reverse end-diastolic flow velocities of the umbilical arteries.

We studied 37 fetuses with absent or reverse end-diastolic flow velocities (AREDFV) of the umbilical arteries with respect to the mode of delivery, fetal acidosis, resistance index of the middle cerebral arteries, and abnormal neurological evaluation at the time of discharge from the department of pediatrics. A control group with normal umbilical artery flow velocity waveforms was matched for gestational age. Fetuses with AREDFV were delivered almost exclusively by cesarean section, in most cases due to fetal distress. Metabolic changes with decreased base excess values were found more often in fetuses with AREDFV, but no difference could be detected with respect to severe fetal acidosis. The number of fetuses with abnormal neurological evaluations during the first 6 months of life was significantly higher in cases with AREDFV than in the control group. More than two thirds of the fetuses with AREDFV showed an abnormal blood flow velocity waveform of the middle cerebral arteries with elevated end-diastolic flow velocities. The combination of extreme premature delivery before 28 weeks of gestation, severe idiopathic respiratory distress syndrome, and a resistance index of the middle cerebral arteries under the 5th percentile seems to be a risk factor for early neurological morbidity in fetuses with AREDFV. Premature delivery of fetuses with AREDFV--as we did during our study in order to prevent intra-uterine death or fetal asphyxia--should therefore be critically regarded in cases under 28 weeks of gestational age and abnormal flow velocity waveforms of the middle cerebral arteries.

Acid-Base Imbalance↗
PubMed1992

Serum electrolytes and capillary blood gases in the management of hypertrophic pyloric stenosis.

The purpose of this study was to assess whether the concentration of serum chloride and other variables, namely serum sodium, potassium and bicarbonate, can be used to predict metabolic acid-base status in infants with hypertrophic pyloric stenosis (HPS) and to assess whether such a prediction is influenced by the state of hydration. One hundred and sixty-three infants with HPS who had at least one set of serum electrolyte and capillary blood gas estimations performed after admission were studied retrospectively. A further 25 infants who had their electrolyte and blood gases measured after at least 12 h of rehydration and correction of the serum sodium and chloride abnormalities were studied prospectively. Stepwise multiple regression analysis, using standard bicarbonate as the dependent variable, revealed serum chloride concentration to be the most powerful independent predictor of standard bicarbonate level (r = -0.69, P less than 0.0001). Other variables did not improve the correlation significantly. In the retrospective study of untreated patients, the calculation of a 90 per cent prediction interval for the model indicated that if the serum chloride level is less than 96 mmol/l, one could be 95 per cent confident (one tail) that the patient was alkalaemic. To be 95 per cent confident that the patient was not alkalaemic, the serum chloride level would have to be greater than 121 mmol/l. The prospective study found that following rehydration a serum chloride level greater than or equal to 106 mmol/l more accurately predicted absence of alkalaemia. We conclude that predictability of acid-base status from measurement of serum chloride depends on the state of hydration of the patient.

Acid-Base Imbalance↗