Jane C. Horrocks

Computer-aided Diagnosis of Acute Abdominal Pain

This paper reports a controlled prospective unselected real-time comparison of human and computer-aided diagnosis in a series of 304 patients suffering from abdominal pain of acute onset. The computing systems overall diagnostic accuracy (91·8%) was significantly higher than that of the most senior member of the clinical team to see each case (79·6%). It is suggested as a result of these studies that the provision of such a system to aid the clinician is both feasible in a real-time clinical setting, and likely to be of practical value, albeit in a small percentage of cases.

Human and Computer-aided Diagnosis of Abdominal Pain: Further Report with Emphasis on Performance of Clinicians

This paper reports a controlled trial of human and computer-aided diagnosis in a series of 552 patients with acute abdominal pain. The overall diagnostic accuracy of the computer-aided system was 91·5% and that of the senior clinician to see each case was 81·2%. However, the clinicians diagnostic performance improved markedly during the period of the trial. The proportion of appendices which perforated before operation fell from 36% to 4% during the trial, and the negative laparotomy rate dropped sharply. After the trial closed in August 1972 these figures reverted towards their pretrial levels. It is suggested that while computer-aided diagnosis is a valuable direct adjunct to the clinician dealing with the “acute abdomen,” he may also benefit in the short-term from the constant feedback he receives and from the disciplines and constraints involved in communicating with the computer.

Clinical findings, early endoscopy, and multivariate analysis in patients bleeding from the upper gastrointestinal tract.

A simple system has been developed to identify patients with upper gastrointestinal tract haemorrhage who run a high risk of continued bleeding or rebleeding. The system is based on six items of patient data available at or soon after arrival in hospital. It was evaluated in a prospective study of 66 patients with upper gastrointestinal tract haemorrhage. Over half of the patients classified by the system into a high-risk category either continued bleeding or rebled after apparent cessation (as against one out of 33 patients in the low-risk category). The high-rish group also had a higher mortality (21%) than those in the low-risk group (nil). The addition or subtraction of early endoscopic findings made little difference to the accuracy of prognosis.

Computer-aided Diagnosis: Description of an Adaptable System, and Operational Experience with 2,034 Cases

This paper describes a system of computer-aided diagnosis using an English Electric KDF9 computer linked to a terminal in a busy clinical department. Data from a series of patients were recorded, coded, and entered into the computer, which then performed a Bayesian analysis and displayed diagnostic probabilities in an adaptable format. Experience in this setting suggests that computer diagnosis may be a valuable aid to the clinician.

Computer-aided diagnosis of “dyspepsia”

Experience with computer-aided diagnosis of “dyspepsia” in a consecutive prospective series of 212 patients coming to surgery is described. Analysis is concentrated upon 122 patients who presented to an outpatient clinic de novo for diagnosis. During their first (outpatient) hospital contact, a firm diagnosis was made in just over half of these patients (though where made, it was usually correct). After full investigation, the diagnostic accuracy (prior to operation) was 92.6%. Using data elicited solely from the house surgeons interviewat the time of admission, the computers overall diagnostic accuracy was 87.7%. The cost of each new computer diagnosis was around 25 new pence (

Clinical Diagnostic Process: An Analysis

0.60). and the time taken was about 5 minutes. In a further small series designed to discriminate between organic and functional dyspepsia, the computer correctly assigned all but 1 of 23 patients with organic disease to the correct disease category. However, almost half of 33 patients with x-ray negative dyspepsia were predicted by the computer to have organic lesions. Time alone will tell whether the computer is a better early predictor of eventual organic disease than currently available radiologic methods.

Observer variation in assessment of results of surgery for peptic ulceration.

An analysis of observations made during 1,307 diagnoses by a total of 28 clinicians (503 diagnoses in real life, and 804 on simulated patients) concerned primarily the interview of patients suffering from abdominal pain. Interviews ranged from 10 to 35 questions, and from “stereotyped” procedures, in which identical (and often irrelevant) questions were asked to each patient, to “adaptive” interviews, in which specific relevant questions were put to each patient. Senior clinicians tended to ask fewer, more relevant questions than their junior counterparts; and urgent cases were dealt with in a more adaptive fashion than routine cases in outpatients. Disappointingly, there was considerable difference between real-life and simulated situations. From these results it is suggested (a) that the “diagnostic process” does not exist, (b) that any automated diagnostic system must be flexible to accommodate the wishes of a variety of clinicians, and (c) that studies based on artificial clinical situations should be treated with extreme caution.

Simulation of Clinical Diagnosis: A Comparative Study

The results of surgery for peptic ulcer may be assessed in many ways. For instance, an assessment may focus on such factors as mortality and morbidity immediately after operation. Early postoperative mortality is, however, low-whatever the operation-and most workers have therefore concentrated on the medium and long-term results of surgery and have dealt with the presence or absence of symptoms attributable to recurrent ulceration or the operative procedure itself, together with the severity of these symptoms. The patients overall status has usually also been graded, the most popular grading system being that devised by Visick.1 As little has been done to assess the reproducibility of these methods of assessment we carried out observer variation studies in a series of 170 patients seen at gastric follow-up clinics in York. We report here our findings.

Production of Artificial “Case Histories” by using a Small Computer

This paper presents a comparison between three different modes of simulation of the diagnostic process—a computer-based system, a verbal mode, and a further mode in which cards were selected from a large board. A total of 34 subjects worked through a series of 444 diagnostic simulations. The verbal mode was found to be most enjoyable and realistic. At the board, considerable amounts of extra irrelevant data were selected. At the computer, the users asked the same questions every time, whether or not they were relevant to the particular diagnosis. They also found the teletype distracting, noisy, and slow. The need for an acceptable simulation system remains, and at present our Minisim and verbal modes are proving useful in training junior clinical students. Future simulators should be flexible, economical, and acceptably realistic—and to us this latter criterion implies the two-way use of speech. We are currently developing and testing such a system.

Use of receiver operating characteristic (ROC) curves to evaluate computer confidence threshold and clinical performance in the diagnosis appendicitis.

This paper describes a method of producing artificial “case histories” by using probability theory and clinical data from a series of 600 patients with acute abdominal pain. A series of 12 such cases were distributed to clinicians, medical students, medical secretaries and technicians, and members of the general public. For each “case” most clinicians concurred with the intended diagnosis. So did the medical secretaries and technicians; indeed this group were more confident of their chosen diagnoses than were the clinicians. It is suggested that clinicians are concerned to a large extent with the consequences of a diagnosis as well as its accuracy, and are motivated to some degree by a fear of the consequences of failure. They may be justified in adopting this policy, for when “errors” in diagnosis are harshly penalized the clinicians were infinitely more effective than any of the other groups.