Thomas L. Griffiths

Learn More
A first step in identifying the content of a document is determining which topics that document addresses. We describe a generative model for documents, introduced by Blei, Ng, and Jordan [Blei, D. M., Ng, A. Y. & Jordan, M. I. (2003) J. Machine Learn. Res. 3, 993-1022], in which each document is generated by choosing a distribution over topics and then(More)
OBJECTIVES To determine the diagnostic accuracy of ureteroscopic biopsy and whether exfoliated cell cytology can improve diagnostic accuracy. METHODS Sixty-two cases of upper tract transitional cell carcinoma were diagnosed by ureteroscopic biopsy and treated by nephroureterectomy. Stage and grade evaluation was possible in 51 cases. Cytology for(More)
OBJECTIVES The aim of this study was to assess the test performance and clinical effectiveness of photodynamic diagnosis (PDD) compared with white light cystoscopy (WLC) in people suspected of new or recurrent bladder cancer. METHODS A systematic review was conducted of randomized controlled trials (RCTs), nonrandomized comparative studies, or diagnostic(More)
The epithelial-mesenchymal transition (EMT) contributes to cancer metastasis. Two ZEB family members, ZEB1 and ZEB2(SIP1), inhibit transcription of the E-cadherin gene and induce EMT in vitro. However, their relevance to human cancer is insufficiently studied. Here, we performed a comparative study of SIP1 and ZEB1 proteins in cancer cell lines and in one(More)
In coming to understand the world-in learning concepts, acquiring language, and grasping causal relations-our minds make inferences that appear to go far beyond the data available. How do we do it? This review describes recent approaches to reverse-engineering human learning and cognitive development and, in parallel, engineering more humanlike machine(More)
CONTEXT Renal cell carcinoma (RCC) accounts for 2-3% of adult malignancies. There remain uncertainties over the oncological outcomes for the surgical management of localised RCC. OBJECTIVE Systematically review relevant literature comparing oncological outcomes of surgical management of localised RCC (T1-2N0M0). EVIDENCE ACQUISITION Relevant databases(More)
We introduce the author-topic model, a generative model for documents that extends Latent Dirichlet Allocation (LDA; Blei, Ng, & Jordan, 2003) to include authorship information. Each author is associated with a multinomial distribution over topics and each topic is associated with a multinomial distribution over words. A document with multiple authors is(More)
CONTEXT For the treatment of localised renal cell carcinoma (RCC), uncertainties remain over the perioperative and quality-of-life (QoL) outcomes for the many different surgical techniques and approaches of nephrectomy. Controversy also remains on whether newer minimally invasive nephron-sparing interventions offer better QoL and perioperative outcomes, and(More)
More than 350,000 new cases of bladder cancer are diagnosed worldwide each year; the vast majority (> 90%) of these are transitional cell carcinomas (TCC). The most important risk factors for the development of bladder cancer are smoking and occupational exposure to toxic chemicals. Painless visible haematuria is the most common presenting symptom of(More)
We address the problem of learning topic hierarchies from data. The model selection problem in this domain is daunting—which of the large collection of possible trees to use? We take a Bayesian approach, generating an appropriate prior via a distribution on partitions that we refer to as the nested Chinese restaurant process. This nonparametric prior allows(More)