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.

Freehand core biopsy in breast cancer: an accurate predictor of tumour grade following neoadjuvant chemotherapy?

Core biopsy is an increasingly used technique in the pre-operative diagnosis of breast carcinoma, as it provides useful prognostic information with respect to tumour type and grade. Neoadjuvant chemotherapy is being used in the treatment of large and locally advanced breast cancers but little is known regarding the correlation between tumour histology on pre-treatment core biopsy and that in residual tumour following primary chemotherapy and surgery. This study aimed to evaluate the accuracy of core biopsy in predicting these features in patients treated with primary chemotherapy. One hundred and thirty-three patients with carcinoma of the breast diagnosed on clinical, radiological and cytological examination underwent core biopsy, followed by primary chemotherapy (with cyclophosphamide, vincristine, doxorubicin and prednisolone) and surgery. The false-negative rate for pre-treatment core biopsy was 14%, with 91% agreement between the grade demonstrated on core biopsy and that in the residual tumour following completion of chemotherapy. Tumour type in the residual post-chemotherapy tumour was predicted by core biopsy in 84%. This study suggests that pre-treatment core biopsy histology accurately predicts residual tumour histology following primary chemotherapy and surgery in patients with breast cancer. (C) 2002 Elsevier Science Ltd. All rights reserved.

The management of radial scars of the breast - does core biopsy help?

Purpose: To compare the effectiveness of fine needle aspiration cytology (FNAC) with core biopsy (CB) in the pre-operative diagnosis of radial scar (RS) of the breast.

Patients and methods: A retrospective analysis was made of all radial scars diagnosed on surgical histology over an 8-year period. Comparison was made between the results of different preoperative needle biopsy techniques and surgical histology findings.

Results: Forty of 47 patients with a preoperative radiological diagnosis of radial scar were included in this analysis. Thirty-eight patients had impalpable lesions diagnosed on mammography and two presented with a palpable lump. FNAC (n=17) was inadequate in 47% of patients, missed two co-existing carcinomas found in this group, and gave a false positive or suspicious result for malignancy in 4 patients. CB (n=23) suggested a RS in 15 patients, but only diagnosed 4 out of 7 co-existing carcinomas found in this group.

Conclusion: CB is more accurate than FNAC in the diagnosis of RS. However, these data demonstrate that CB may offer little to assist in the management of patients with RS. In summary, this paper advocates the use of CB in any lesion with a radiological suspicion of carcinoma and diagnostic excision of all lesions thought to be typical of RS on mammography.

Questions of 'h' in Northern Ireland: Breathing New Life on the Aspiration Theory

This paper considers existing ideas concerning pronunciation of the letter name for (LNH) in Northern Irish English (NIE). Traditionally, the status of LNH realisation as an ethnic marker has gone unquestioned: Catholics are thought to say [het&Mac186;] while the Protestant norm is assumed to be [etS]. The phonetic difference between these realisations is consistently described as word-initial aspiration versus non-aspiration, with aspiration attributed exclusively to Irish language influence. Here, we show that an explanation based on aspiration alone is phonologically unsatisfying and question whether aspiration is, in fact, an Irish language or ethnically dictated phenomenon. It is further suggested here that the overwhelming stigmatisation of LNH realisation may be responsible for blocking a potential sound change in NIE. While this paper is not intended as a detailed account of ethnolinguistic differences in NI phonology, it engages critically with the over-simplistic and widespread notion that LNH realisation is a result of transfer from the Irish language to the English used by Catholics in Northern Ireland.

Necrospermia and chronic spinal cord injury.

OBJECTIVE: To determine whether improvement in quality of semen over 4 consecutive days of electroejaculation in men with chronic spinal cord injury (SCI) was consistent with epididymal necrospermia. DESIGN: Prospective study of a random sample of men with SCI. SETTING: A southeastern Australian SCI management center in collaboration with the specialist andrology service of a university-based department of obstetrics and gynecology in a tertiary referral hospital. PATIENT(S): Nine men with chronic spinal cord injury. INTERVENTION(S): Semen samples were obtained by using electroejaculation, and testicular biopsy samples were obtained by using fine-needle tissue aspiration. MAIN OUTCOME MEASURE(S): Semen analysis was performed according to World Health Organization criteria. Testicular biopsy and electron microscopy were done by using standard techniques. RESULT(S): During up to 4 days of consecutive-day electroejaculation, sperm motility and viability in semen obtained from men with chronic SCI increased by an average of 23% on days 2 and 3. The severity of the degenerative changes and the numbers of spermatozoa affected on day 1 became less marked by day 4. The changes were not present in late spermatids obtained from testicular biopsies. CONCLUSION(S): The asthenospermia of chronic SCI is similar to epididymal necrospermia and can be improved by consecutive-day electroejaculation.