Dean Billheimer

Dean Billheimer

Professor, Public Health
Director, Statistical Consulting
Professor, Statistics-GIDP
Professor, BIO5 Institute
Member of the General Faculty
Member of the Graduate Faculty
Primary Department
Contact
(520) 626-9902

Work Summary

My research develops new clinical trial and experimental study designs to allow 'learning from data' more efficiently. My research also develops new analysis methods to understand latent structure in data. This allows better understanding of disease processes, better targeting of existing treatments, and development of more effective new treatments. Finally, I am developing new statistical methods based on prediction of future events.

Research Interest

Dean Billheimer, PhD, works with the Arizona Statistics Consulting Laboratory (StatLab) to partner with scientists and physicians to advance discovery and understanding. The 'Stat Lab' provides statistical expertise, personnel and computing resources to facilitate study design and conduct, data acquisition protocols, data analysis, and the preparation of grants and manuscripts. Dr. Billheimer also works to adapt and develop new statistical methods to address emerging problems in science and medicine. Dr. Billheimer facilitates discovery translation and economic development by consulting with public and private organizations external to the University of Arizona. Keywords: Biostatistics, Bioinformatics, Study Design, Bayesian Analysis

Publications

Chen, S., Ming, L. i., Hong, D., Billheimer, D., Huiming, L. i., Xu, B. J., & Shyr, Y. (2009). A novel comprehensive wave-form MS data processing method. Bioinformatics, 25(6), 808-814.

PMID: 19176559;PMCID: PMC2732299;Abstract:

Motivation: Mass spectrometry (MS) can generate high-throughput protein profiles for biomedical research to discover biologically related protein patterns-biomarkers. The noisy functional MS data collected by current technologies, however, require consistent, sensitive and robust data-processing techniques for successful biomedical application. Therefore, it is important to detect features precisely for each spectrum, quantify them well and assign a unique label to features from the same protein-peptide across spectra. Results: In this article, we propose a new comprehensive MS data preprocessing package, Wave-spec, which includes several novel algorithms. It can overcome several conventional difficulties. Wave-spec can be applied to multiple types of MS data generated with different MS technologies. Results from this new package were evaluated and compared to several existing approaches based on a MALDI-TOF MS dataset. © The Author 2009. Published by Oxford University Press. All rights reserved.

Goldstein, R. E., Billheimer, D., Martin, W. H., Richards, K., Jacobs, J. K., Udelsman, R., & L., G. (2003). Sestamibi Scanning and Minimally Invasive Radioguided Parathyroidectomy Without Intraoperative Parathyroid Hormone Measurement. Annals of Surgery, 237(5), 722-731.

PMID: 12724639;PMCID: PMC1514518;Abstract:

Objective: To evaluate the results of a large series of patients undergoing minimally invasive radioguided parathyroidectomy (MIRP) in which routine use of the intraoperative parathyroid hormone assay was not used, and to investigate characteristics between patients who had positive preoperative parathyroid scans versus those with negative scans. Summary Background Data: The technique of parathyroidectomy has traditionally involved bilateral exploration of the neck under general endotracheal anesthesia. Parathyroid imaging using technetium-99m sestamibi (MIBI) has evolved and can localize the adenomas in 80% to 90% of patients. The MIRP technique combines parathyroid scintigraphy with a hand-held gamma detector used intraoperatively to guide the surgeon to the adenoma in patients with positive MIBI scans. Central to this technique or other unilateral approaches is a positive MIBI scan. Methods: One hundred seventy-three atients with primary hyperparathyroidism operated on by a single surgeon between January 1998 and July 2002 were included. One hundred twelve patients underwent the MIRP procedure and by definition had a positive preoperative parathyroid scan. The technique involved injecting 20 mCi MIBI 1 hour before the surgical procedure in patients who preoperatively had positive MIBI imaging. Patients had the choice of general or MAC anesthesia. Using an incision of less than 4 cm, the dissection to the adenoma was guided by the Navigator 11-mm probe. These 112 patients and 4 additional patients who for various reasons did not have the MIRP procedure yet had positive MIBI scans were compared to 57 patients who had clearly negative MIBI parathyroid imaging. Results: Follow-up data were available for 108 of 112 patients who underwent MIRP. No patients had persistent hypercalcemia. The long-term success rate for the MIRP group was 98%. Fifty-two percent of the MIRP procedures were performed using MAC anesthesia. Overall, gland weight and serum PTH were related to the probability of a positive MIBI scan. Multiple logistic regression revealed that females were more likely to exhibit positive scans than were males for any fixed serum PTH level. For females, there was a significant relationship between increasing serum parathyroid hormone and a positive MIBI scan. Conversely, in males, the relationship between scan positivity and serum parathyroid hormone was weaker. Conclusions: The MIRP technique without routine intraoperative serum parathyroid hormone measurement resulted in an excellent cure rate for primary hyperparathyroidism. As the MIRP technique as well as other techniques for unilateral cervical exploration are predicated on a positive parathyroid scan, the possible effect of gender on the sensitivity of MIBI scintigraphy for the detection of parathyroid adenomas warrants further investigation.

Lieber, C. A., Majumder, S. K., Billheimer, D., Ellis, D. L., & Mahadevan-Jansen, A. (2008). Raman microspectroscopy for skin cancer detection in vitro. Journal of Biomedical Optics, 13(2).

PMID: 18465976;Abstract:

We investigate the potential of near-infrared Raman microspectroscopy to differentiate between normal and malignant skin lesions. Thirty-nine skin tissue samples consisting of normal, basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma from 39 patients were investigated. Raman spectra were recorded at the surface and at 20-μm intervals below the surface for each sample, down to a depth of at least 100μm. Data reduction algorithms based on the nonlinear maximum representation and discrimination feature (MRDF) and discriminant algorithms using sparse multinomial logistic regression (SMLR) were developed for classification of the Raman spectra relative to histopathology. The tissue Raman spectra were classified into pathological states with a maximal overall sensitivity and specificity for disease of 100%. These results indicate the potential of using Raman microspectroscopy for skin cancer detection and provide a clear rationale for future clinical studies. © 2008 Society of Photo-Optical Instrumentation Engineers.

Fowler, K., Poehling, K., Billheimer, D., Hamilton, R., Huiyun, W. u., Mulder, J., & Frangoul, H. (2006). Hospice referral practices for children with cancer: A survey of pediatric oncologists. Journal of Clinical Oncology, 24(7), 1099-1104.

PMID: 16505429;Abstract:

Purpose: To examine hospice referral patterns among pediatric oncologists and identify barriers to referral. Methods: A self-administered survey was sent to 1,200 pediatric oncologists who are members of Children's Oncology Group. Two electronic mail messages followed by traditional mail surveys were sent to eligible physicians. Pediatricians and pediatric oncologists developed, pretested, and modified the survey for item clarification. Results: Of 944 eligible pediatric oncologists surveyed, 632 replied, yielding a response rate of 67%. Most respondents reported having access to palliative care programs (65%) and hospice services (85%), but few (27%) had access to inpatient hospice services. More respondents reported feeling comfortable managing end-of-life pain than psychological issues (86% v 67%, respectively). Many pediatric oncologists (62%) reported that half or more of their patients died in the hospital. In multivariate analysis, physicians with access to hospice that accepts patients receiving chemotherapy had more patients die at home than in hospital compared with physicians without access to such services (P = .007). The probability of hospice referral was positively associated with the presence of a hospice facility (P .001) and with a larger size oncology group (P = .024). Only 2.5% of respondents referred patients at the time of relapse. Continued therapy was cited as the most common reason for not making a referral, and was significantly higher when hospice did not admit children receiving chemotherapy (P = .002). Conclusion: Hospice referral for children with cancer is usually made late in the course of their disease and might improve if hospice admits patients who are actively receiving chemotherapy. © 2006 by American Society of Clinical Oncology.

M'Koma, A. E., Blum, D. L., Norris, J. L., Koyama, T., Billheimer, D., Motley, S., Ghiassi, M., Ferdowsi, N., Bhowmick, I., Chang, S. S., Fowke, J. H., Caprioli, R. M., & Bhowmick, N. A. (2007). Detection of pre-neoplastic and neoplastic prostate disease by MADI profiling of urine. Biochemical and Biophysical Research Communications, 353(3), 829-834.

PMID: 17194448;PMCID: PMC2562600;Abstract:

The heterogeneous progression to the development of prostate cancer (PCa) has precluded effective early detection screens. Existing prostate cancer screening paradigms have relatively poor specificity for cancer relative to other prostate diseases, commonly benign prostatic hyperplasia (BPH). A method for discrimination of BPH, HGPIN, and PCa urine proteome was developed through testing 407 patient samples using matrix assisted laser desorption-mass spectrometry time of flight (MALDI-TOF). Urine samples were adsorbed to reverse phase resin, washed, and the eluant spotted directly for MALDI-TOF analysis of peptides. The processing resolved over 130 verifiable signals of a mass range of 1000-5000 m/z to suggest 71.2% specificity and 67.4% sensitivity in discriminating PCa vs. BPH. Comparing BPH and HGPIN resulted in 73.6% specificity and 69.2% sensitivity. Comparing PCa and HGPIN resulted in 80.8% specificity and 81.0% sensitivity. The high throughput, low-cost assay method developed is amenable for large patient numbers required for supporting biomarker identification. © 2006.