EXPERT-SYSTEM SUPPORT USING BAYESIAN BELIEF NETWORKS IN THE DIAGNOSIS OF FINE-NEEDLE ASPIRATION BIOPSY SPECIMENS OF THE BREAST

Aim-To develop an expert system model for the diagnosis of fine needle aspiration cytology (FNAC) of the breast.

Methods-Knowledge and uncertainty were represented in the form of a Bayesian belief network which permitted the combination of diagnostic evidence in a cumulative manner and provided a final probability for the possible diagnostic outcomes. The network comprised 10 cytological features (evidence nodes), each independently linked to the diagnosis (decision node) by a conditional probability matrix. The system was designed to be interactive in that the cytopathologist entered evidence into the network in the form of likelihood ratios for the outcomes at each evidence node.

Results-The efficiency of the network was tested on a series of 40 breast FNAC specimens. The highest diagnostic probability provided by the network agreed with the cytopathologists' diagnosis in 100% of cases for the assessment of discrete, benign, and malignant aspirates. A typical probably benign cases were given probabilities in favour of a benign diagnosis. Suspicious cases tended to have similar probabilities for both diagnostic outcomes and so, correctly, could not be assigned as benign or malignant. A closer examination of cumulative belief graphs for the diagnostic sequence of each case provided insight into the diagnostic process, and quantitative data which improved the identification of suspicious cases.

Conclusion-The further development of such a system will have three important roles in breast cytodiagnosis: (1) to aid the cytologist in making a more consistent and objective diagnosis; (2) to provide a teaching tool on breast cytological diagnosis for the non-expert; and (3) it is the first stage in the development of a system capable of automated diagnosis through the use of expert system machine vision.

Refining the diagnosis and EGFR status of non-small cell lung carcinoma in biopsy and cytologic material, using a panel of mucin staining, TTF-1, cytokeratin 5/6, and P63, and EGFR mutation analysis.

INTRODUCTION: The dichotomization of non-small cell carcinoma (NSCLC) subtype into squamous (SQCC) and adenocarcinoma (ADC) has become important in recent years and is increasingly required with regard to management. The aim of this study was to determine the utility of a panel of commercially available antibodies in refining the diagnosis on small biopsies and also to determine whether cytologic material is suitable for somatic EGFR genotyping in a prospectively analyzed series of patients undergoing investigation for suspected lung cancer. METHODS: Thirty-two consecutive cases of NSCLC were first tested using a panel comprising cytokeratin 5/6, P63, thyroid transcription factor-1, 34betaE12, and a D-PAS stain for mucin, to determine their value in refining diagnosis of NSCLC. After this test phase, two further pathologists independently reviewed the cases using a refined panel that excluded 34betaE12 because of its low specificity for SQCC, and refinement of diagnosis and concordance were assessed. Ten cases of ADC, including eight derived from cytologic samples, were sent for EGFR mutation analysis. RESULTS: There was refinement of diagnosis in 65% of cases of NSCLC to either SQCC or ADC in the test phase. This included 10 of 13 cases where cell pellets had been prepared from transbronchial needle aspirates. Validation by two further pathologists with varying expertise in lung pathology confirmed increased refinement and concordance of diagnosis. All samples were adequate for analysis, and they all showed a wild-type EGFR genotype. CONCLUSION: A panel comprising cytokeratin 5/6, P63, thyroid transcription factor-1, and a D-PAS stain for mucin increases diagnostic accuracy and agreement between pathologists when faced with refining a diagnosis of NSCLC to SQCC or ADC. These small samples, even cell pellets derived from transbronchial needle aspirates, seem to be adequate for EGFR mutation analysis.