A range of studies have proven the feasibility of TM for extracting structured information from clinical narratives such as those found in pathology or radiology reports. In this article, we provide a critical overview of the current state of the art for TM related to cancer. The review highlighted a strong bias towards symbolic methods, e.g. named entity recognition (NER) based on dictionary lookup and information extraction (IE) relying on pattern matching. The F-measure of NER ranges between 80% and 90%, while that of IE for simple tasks is in the high 90s. To further improve the performance, TM approaches need to deal effectively with idiosyncrasies of the clinical sublanguage such as non-standard abbreviations as well as a high degree of spelling and grammatical errors. This requires a shift from rule-based methods to machine learning following the success of similar trends in biological applications of TM. Machine learning approaches require large training datasets, but clinical narratives are not readily available for TM research due to privacy and confidentiality concerns. This issue remains the main bottleneck for progress in this area. In addition, there is a need for a comprehensive cancer ontology that would enable semantic representation of textual information found in narrative reports.

Concurrent gemcitabine-based chemoradiotherapy (ie, GemX) produces a high response rate in MIBC and has durable local control and acceptable toxicity, which allows patients to preserve their own bladder. This treatment modality warrants additional investigation in a phase III setting.

No abstract

No abstract

Objective: Day-to-day anatomical variations complicate bladder cancer radiotherapy treatment. This work quantifies the impact on target coverage and irradiated normal tissue volume for different adaptive strategies. Methods: 20 patients were retrospectively planned using different three-dimensional conformal radiotherapy treatment strategies for whole-bladder carcinoma: (i) ''conventional'' treatment used isotropic expansion of the clinical target volume (CTV) by 15 mm to the planning target volume (PTV) for daily treatment; (ii) ''plan of the day'' used daily volumetric on-treatment imaging [cone beam CT (CBCT)] to select from four available plans with varying superior PTV margins; (iii) ''composite'' strategies used ontreatment CBCTs from Fractions 1-3 to inform a composite CTV and adapted PTV (5-and 10-mm margins for composite 1 and composite 2, respectively) for subsequent treatment. Target coverage was evaluated from available CBCTs (the first three fractions then the minimum weekly thereafter), and the reduction in the irradiated volume (i.e. within the 95% isodose) was quantified. Results: Plan of the day improved target coverage (i.e. all of the bladder within the 95% isodose throughout the treatment) relative to conventional treatment (p50.10), while no such benefit was observed with composite 2. Target coverage was reduced with composite 1 relative to conventional treatment. The mean irradiated volume was reduced by 17.2%, 35.0% and 14.6% relative to conventional treatment, for plan of the day, composite 1 and composite 2, respectively (p,0.01 in all cases). No parameters predictive of large changes in bladder volume later in the treatment were identified. Conclusions: Adaptive techniques can maintain or improve target coverage while allowing for reduced irradiated volume and possibly reduced toxicity. The plan-of-theday technique appeared to provide the optimal balance between target coverage and normal tissue sparing. Advances in knowledge: This study suggests that plan-of-the-day techniques will provide optimal outcomes for adaptive bladder radiotherapy.