Harry B. Burke

Artificial neural networks improve the accuracy of cancer survival prediction

The TNM staging system originated as a response to the need for an accurate, consistent, universal cancer outcome prediction system. Since the TNM staging system was introduced in the 1950s, new prognostic factors have been identified and new methods for integrating prognostic factors have been developed. This study compares the prediction accuracy of the TNM staging system with that of artificial neural network statistical models.

Group Consensus Reports from the Consensus Conference on Focal Treatment of Prostatic Carcinoma, Celebration, Florida, February 24, 2006

( EPORT OF CONSENSUS GROUP 1: ATHOBIOLOGY OF PROSTATE CANCER: MPLICATIONS FOR FOCAL THERAPY ocal ablative therapy may be reasonable for some atients with prostate cancer; selection factors include variety of clinical and pathologic factors in combiation with informed patient choice. Our group evalated 4 specific pathologic features that may influence his treatment decision. We reviewed the published iterature for applicable studies regarding the natural istory of prostate cancer, multifocality, cancer volme, and accuracy of cancer detection by current ethods. Results were as follows:

Predicting response to adjuvant and radiation therapy in patients with early stage breast carcinoma

Screening and surveillance is increasing the detection of early stage breast carcinoma. The ability to predict accurately the response to adjuvant therapy (chemotherapy or tamoxifen therapy) or postlumpectomy radiation therapy in these patients can be vital to their survival, because this prediction determines the best postsurgical therapy for each patient.

Electronic health records improve clinical note quality

Background and objective The clinical note documents the clinicians information collection, problem assessment, clinical management, and its used for administrative purposes. Electronic health records (EHRs) are being implemented in clinical practices throughout the USA yet it is not known whether they improve the quality of clinical notes. The goal in this study was to determine if EHRs improve the quality of outpatient clinical notes. Materials and methods A five and a half year longitudinal retrospective multicenter quantitative study comparing the quality of handwritten and electronic outpatient clinical visit notes for 100 patients with type 2 diabetes at three time points: 6 months prior to the introduction of the EHR (before-EHR), 6 months after the introduction of the EHR (after-EHR), and 5 years after the introduction of the EHR (5-year-EHR). QNOTE, a validated quantitative instrument, was used to assess the quality of outpatient clinical notes. Its scores can range from a low of 0 to a high of 100. Sixteen primary care physicians with active practices used QNOTE to determine the quality of the 300 patient notes. Results The before-EHR, after-EHR, and 5-year-EHR grand mean scores (SD) were 52.0 (18.4), 61.2 (16.3), and 80.4 (8.9), respectively, and the change in scores for before-EHR to after-EHR and before-EHR to 5-year-EHR were 18% (p<0.0001) and 55% (p<0.0001), respectively. All the element and grand mean quality scores significantly improved over the 5-year time interval. Conclusions The EHR significantly improved the overall quality of the outpatient clinical note and the quality of all its elements, including the core and non-core elements. To our knowledge, this is the first study to demonstrate that the EHR significantly improves the quality of clinical notes.

Radiation Exposure and Cost Influence Physician Medical Image Decision Making A Randomized Controlled Trial

Background: It is estimated that 20%–40% of advanced medical imaging in the United States is unnecessary, resulting in patient overexposure to radiation and increasing the cost of care. Previous imaging utilization studies have focused on clinical appropriateness. An important contributor to excessive use of advanced imaging may be a physician “knowledge gap” regarding the safety and cost of the tests. Objectives: To determine whether safety and cost information will change physician medical image decision making. Research Design: Double-blinded, randomized controlled trial. Following standardized case presentation, physicians made an initial imaging choice. This was followed by the presentation of guidelines, radiation exposure and health risk, and cost information. Results: Approximately half (57 of 112, 50.9%) of participants initially selected computed tomography (CT). When presented with guideline recommendations, participants did not modify their initial imaging choice (P=0.197). A significant reduction (56.3%, P<0.001) in CT ordering occurred after presentation of radiation exposure/health risk information; ordering changed to magnetic resonance imaging or ultrasound (US). A significant reduction (48.3%, P<0.001) in CT and magnetic resonance imaging ordering occurred after presentation of Medicare reimbursement information; ordering changed to US. The majority of physicians (31 of 40, 77.5%) selecting US never modified their ordering. No significant relationship between physician demographics and decision making was observed. Conclusions: This study suggests that physician decision making can be influenced by safety and cost information and the order in which information is provided to physicians can affect their decisions.

Clinical and laboratory evaluation of all-trans retinoic acid modulation of chemotherapy in patients with acute myelogenous leukaemia.

All‐trans retinoic acid (ATRA) is synergistic with chemotherapy in leukaemia cell lines. We treated 53 patients with newly diagnosed acute myelogenous leukaemia (AML) with high‐dose cytarabine‐based chemotherapy followed by ATRA. Peripheral blood and bone marrow samples were obtained to study the effect of in vitro exposure to ATRA and to measure apoptosis and bcl‐2. The response rate was 72% for patients under age 60 years and 46% for patients aged 60 years or above. There was no difference in the percentage of responding patients, time to recurrence or overall survival for patients receiving chemotherapy with ATRA vs. historical controls receiving chemotherapy without ATRA. After in vitro exposure of day 3 bone marrow samples to ATRA, there was an increase in apoptotic cells in 25% of patient samples compared with samples not exposed to ATRA. Later date of peak apoptosis in peripheral blood and higher percentage of apoptotic cells in bone marrow on day 3 of treatment were associated with lack of clinical response to treatment. Increased bcl‐2 in patient samples was associated with shorter time to recurrence and poor cytogenetic risk. The addition of ATRA to chemotherapy did not improve patient outcome. However, evidence of in vitro response to ATRA in 25% of patients suggests that retinoid pathways should be studied further in patients with AML.

QNOTE: an instrument for measuring the quality of EHR clinical notes

Background and objective The outpatient clinical note documents the clinicians information collection, problem assessment, and patient management, yet there is currently no validated instrument to measure the quality of the electronic clinical note. This study evaluated the validity of the QNOTE instrument, which assesses 12 elements in the clinical note, for measuring the quality of clinical notes. It also compared its performance with a global instrument that assesses the clinical note as a whole. Materials and methods Retrospective multicenter blinded study of the clinical notes of 100 outpatients with type 2 diabetes mellitus who had been seen in clinic on at least three occasions. The 300 notes were rated by eight general internal medicine and eight family medicine practicing physicians. The QNOTE instrument scored the quality of the note as the sum of a set of 12 note element scores, and its inter-rater agreement was measured by the intraclass correlation coefficient. The Global instrument scored the note in its entirety, and its inter-rater agreement was measured by the Fleiss κ. Results The overall QNOTE inter-rater agreement was 0.82 (CI 0.80 to 0.84), and its note quality score was 65 (CI 64 to 66). The Global inter-rater agreement was 0.24 (CI 0.19 to 0.29), and its note quality score was 52 (CI 49 to 55). The QNOTE quality scores were consistent, and the overall QNOTE score was significantly higher than the overall Global score (p=0.04). Conclusions We found the QNOTE to be a valid instrument for evaluating the quality of electronic clinical notes, and its performance was superior to that of the Global instrument.

Histologic grade as a prognostic factor in breast carcinoma

Although I welcome efforts to go beyond the current TNM practice of considering only anatomic extent variables in the assessment of the prognosis of breast carcinoma, Drs. Burke and Henson fail to appropriately address the practical clinical problem of the potential use of grade as a prognostic indicator and guide to treatment within the TNM framework. That tumor size and lymph node status outweigh grade alone is not in question, nor is it at all surprising that combining size and grade, as they have done, demonstrates little additional statistical effect for the inclusion of grade. However, that information is irrelevant to the problem of selecting treatment for individual patients, because it makes no attempt to stratify by tumor size and then examine grade for each T classification. Virtually all of the material I have presented supports the position that grade is a usable prognostic indicator in this clinical context. Furthermore, this position has been validated by Dr. Henson’s own findings in previously published analyses of Surveillance, Epidemiology, and End Results (SEER) data. Carter, with Dr. Henson as a coauthor, examined SEER cases for the relation between tumor size and lymph node status and noted ‘‘Within the group, however, are subsets that have increasing or decreasing survival experiences for the same tumor size and node status.’’ In a subsequent report, Henson et al. clearly demonstrated that within each T category, grade is related to survival, for both lymph node negative and positive patients. TNM system Stage IV cases also were analyzed. Contrary to the editorial, these had been graded in approximately the same proportion as the entire group, 27.0% as opposed to the overall 24.8%. The 5and 10-year prognostic indices of Henson et al. demonstrated a consistent survival advantage for patients with Grade 1 disease, including those patients with Stage IV disease. They stated ‘‘Histologic grade, when used in conjunction with stage of disease, can improve the prediction of outcome. Our results also indicate that a prognostic index can be created for breast cancer using a combination of stage of disease and histologic grade.’’ The current editorial analysis finds that grade adds nothing to See editorial counterpoint on pages 1703–5, the predictive value of new molecular genetic factors. Omitted is any and referenced original article on pages 1708– statement of whether the reverse is true. It remains unclear what, if 16, this issue. anything, the new factors add to the prognostic value of grade. The TNM system is simple to use and easy to understand. It is Address for reprints: Nicholas E. Roberti, M.D., an excellent tool for subset identification and for communication 58 St. Malo Beach, Oceanside, CA 92054. among clinicians, and is a great aid in patient education. The TNM Received June 27, 1997; accepted July 9, 1997. nomenclature has proven its clinical value and will not be discarded

Statistical Analysis of Complex Systems in Biomedicine

The future explanatory power in biomedicine will be at the molecular-genetic level of analysis (rather than the epidemiologic-demographic or anatomic-cellular levels). This is the level of complex systems. Complex systems are characterized by nonlinearity and complex interactions. It is difficult for traditional statistical methods to capture complex systems because traditional methods attempt to find the model that best fits the statistician’s understanding of the phenomenon; complex systems are difficult to understand and therefore difficult to fit with a simple model. Artificial neural networks are nonparametric regression models. They can capture any phenomena, to any degree of accuracy (depending on the adequacy of the data and the power of the predictors), without prior knowledge of the phenomena. Further, artificial neural networks can be represented, not only as formulae, but also as graphical models. Graphical models can increase analytic power and flexibility. Artificial neural networks are a powerful method for capturing complex phenomena, but their use requires a paradigm shift, from exploratory analysis of the data to exploratory analysis of the model.

High-resolution transrectal ultrasound: pilot study of a novel technique for imaging clinically localized prostate cancer.

OBJECTIVES To determine how high-resolution transrectal ultrasound (HiTRUS) compares with conventional TRUS (LoTRUS) for the visualization of prostate cancer. METHODS AND MATERIALS Twenty-five men with known prostate cancer scheduled for radical prostatectomy were preoperatively imaged with both LoTRUS (5MHz) and HiTRUS (21MHz). Dynamic cine loops and still images for each modality were saved and subjected to blinded review by a radiologist looking for hypoechoic foci ≥ 5 mm in each sextant of the prostate. Following prostatectomy, areas of prostate cancer ≥ 5 mm on pathologic review were anatomically correlated to LoTRUS and HiTRUS findings. The accuracy of LoTRUS and HiTRUS to visualize prostate cancer in each sextant of the prostate and to identify high-grade and locally advanced disease was assessed. The McNemar test was used to compare sensitivity and specificity and paired dichotomous outcomes between imaging modalities. RESULTS Among 69 sextants with pathologically identified cancerous foci at radical prostatecomy, HiTRUS visualized 45 and missed 24, whereas LoTRUS visualized 26 and missed 43. Compared with LoTRUS, HiTRUS demonstrated improved sensitivity (65.2% vs. 37.7%) and specificity (71.6% vs. 65.4%). HiTRUSs agreement with pathologic findings was twice as high as LoTRUS (P = 0.006). HiTRUS provided a nonsignificant increase in visualization of high-grade lesions (84% vs. 60%, P = 0.11). CONCLUSIONS HiTRUS appears promising for prostate cancer imaging. Our initial experience suggests superiority to LoTRUS for the visualization of cancerous foci, and supports proceeding with a clinical trial in the biopsy setting.