Using automatically extracted information from mammography reports for decision-support

OBJECTIVE

To evaluate a system we developed that connects natural language processing (NLP) for information extraction from narrative text mammography reports with a Bayesian network for decision-support about breast cancer diagnosis. The ultimate goal of this system is to provide decision support as part of the workflow of producing the radiology report.

MATERIALS AND METHODS

We built a system that uses an NLP information extraction system (which extract BI-RADS descriptors and clinical information from mammography reports) to provide the necessary inputs to a Bayesian network (BN) decision support system (DSS) that estimates lesion malignancy from BI-RADS descriptors. We used this integrated system to predict diagnosis of breast cancer from radiology text reports and evaluated it with a reference standard of 300 mammography reports. We collected two different outputs from the DSS: (1) the probability of malignancy and (2) the BI-RADS final assessment category. Since NLP may produce imperfect inputs to the DSS, we compared the difference between using perfect ("reference standard") structured inputs to the DSS ("RS-DSS") vs NLP-derived inputs ("NLP-DSS") on the output of the DSS using the concordance correlation coefficient. We measured the classification accuracy of the BI-RADS final assessment category when using NLP-DSS, compared with the ground truth category established by the radiologist.

RESULTS

The NLP-DSS and RS-DSS had closely matched probabilities, with a mean paired difference of 0.004±0.025. The concordance correlation of these paired measures was 0.95. The accuracy of the NLP-DSS to predict the correct BI-RADS final assessment category was 97.58%.

CONCLUSION

The accuracy of the information extracted from mammography reports using the NLP system was sufficient to provide accurate DSS results. We believe our system could ultimately reduce the variation in practice in mammography related to assessment of malignant lesions and improve management decisions.

Comparison of methods for prediction of prostate cancer in Turkish men with PSA levels of 0-10 ng/mL

PURPOSE

Several concepts to improve the diagnostic accuracy of prostate specific antigen (PSA) for prediction of prostate cancer have been studied. The aim of this study was to examine and compare the methods used for improving the diagnostic accuracy of PSA in a country with low incidence of prostate cancer.

METHODS

997 patients with prostate biopsy were included into study. Predictive models using PSA, PSA density (PSAD), free PSA/total PSA (f/tPSA), binary logistic regression (LR) analysis, artificial neural networks (ANNs), and decision trees (DTs) have been developed. For LR, ANNs and DTs, a validation group consisting of 241 cases was reserved.

RESULTS

193 (19%) biopsies out of 997 showed prostatic cancer. Median PSAD in patients with malignant and benign lesions were 0.21 and 0.16, respectively (p<0.001). According to 25% f/tPSA cut-off level, 18.4% of the patients with PSA<25% and 16.0% of the patients with PSA>25% had prostate cancer (p=0.423). Receiver operating characteristics (ROC) area under the curve (AUC) values for PSA, PSA density, f/tPSA, LR, ANNs, and DTs were 0.587, 0.625, 0.560, 0.678, 0.644, and 0.698, respectively. ROC AUCs in the validation group for LR, ANNs and DTs were 0.717, 0.516 and 0.629 respectively.

CONCLUSIONS

For cases with f/tPSA<25%, no increased probability for prostatic carcinoma was observed. Multivariate models have higher AUCs than PSA, PSAD or f/tPSA. LR, DTs and ANNs showed similar results, however application of ANNs to the validation group produced a significantly lower AUC, limiting the value of ANNs in this situation.

Relationship between mast cell and iNOS expression in testicular tissue associated with infertility

The objective of this study was to investigate mast cells and iNOS expression in testis tissue, and to correlate these results with spermatogenic disorders. A total of 136 testicular biopsies were obtained from the testes of 80 patients with infertility. Their age ranged from 21 to 45 years. The biopsy specimens were immunohistochemically stained with antihuman tryptase for mast cells. In each section, all interstitial fields were evaluated for the total number of mast cells as well as the total number of Leydig cells. The number of mast cells per Leydig cell was calculated and recorded as mast cell index. Immunohistochemical iNOS staining was evaluated semiquantitatively according to intensity and the proportion of the stained cells. There was a significant increase of the mast cell index in all groups with testicular disorder compared with normal spermatogenesis group (p < 0.05). Increase of the index was in the order of hypospermatogenesis, maturation arrest and SCO, and index of SCO group was especially higher, i.e, more than twice than other groups. iNOS score was significantly higher in the SCO group than in the men with normal spermatogenesis, hypospermatogenesis, and maturation arrest (p < 0.05). Finally, a significantly statistical correlation was found between the iNOS score and mast cells index (r = 0.758, p = 0.001). Increase of mast cell index was observed in the groups of infertile testis, and high expression of iNOS in Leydig cells was associated with the highest mast cell index in SCO, the lesion with the most severe damage of the germ cell.

The assessment of PTEN tumor suppressor gene in combination with Gleason scoring and serum PSA to evaluate progression of prostate carcinoma

OBJECTIVE

The purpose of the study was to determine if the tumor suppressor gene phosphate and tensin homolog (PTEN) (mutated in multiple advanced cancers 1) in combination with Gleason scoring and serum prostate specific antigen (PSA) could be employed to better predict the progression of prostate carcinoma.

MATERIALS AND METHODS

The study group consisted of 43 patients with benign prostate hyperplasia (BPH), 15 with organ confined prostate carcinoma (OCPCa), and 18 with advanced prostate carcinoma (APCa). Prostate tissue samples were obtained from radical prostatectomy, transurethral resection, and TRUS guided trans-rectal needle biopsy and then evaluated for biomarker expression. The clinical stage was assessed according to tumor node metastasis classification and grade according to Gleason system. Serum PSA was measured by conventional techniques and Western blotting analysis was used to determine PTEN expression in the primary tissue. Multivariate analysis was performed to analyze whether these markers could individually predict the progression of prostate carcinoma.

RESULTS

APCa patients displayed higher Gleason scores and serum PSA levels. But much lower PTEN expression was detected in prostate of APCa patients compared to patients with BPH or OCPCa. Hormone refractory (HR) and hormone sensitive (HS) APCa cases did not yield any significant differences in terms of Gleason scoring, serum PSA and PTEN expression. PSA levels were significantly higher in patients with OCPCa or APCa compared to patients with BPH.

CONCLUSION

Our results suggested that both PTEN and serum PSA appeared to be useful as independent markers to depict the nature of tumor behavior as benign or malign. In addition, PTEN also appeared to be useful as an independent marker to predict the progression of prostate carcinoma.