Search

Supplemental material, sj-pdf-2-smm-10.1177_0962280220952833 for Optimal two-stage sampling for mean estimation in multilevel populations when cluster size is informative by Francesco Innocenti, Math JJM Candel, Frans ES Tan and Gerard JP van Breukelen in Statistical Methods in Medical Research

Supplemental material, sj-pdf-1-smm-10.1177_0962280220952833 for Optimal two-stage sampling for mean estimation in multilevel populations when cluster size is informative by Francesco Innocenti, Math JJM Candel, Frans ES Tan and Gerard JP van Breukelen in Statistical Methods in Medical Research

Over the last five years deep learning has progressed tremendously in both image recognition and natural language processing. Now it is increasingly applied to other data rich fields. In drug discovery, recurrent neural networks (RNNs) have been shown to be an effective method to generate novel chemical structures in the form of SMILES. However, ligands generated by current methods used to provide relatively little diversity and do not fully cover the whole chemical space occupied by known ligands. Here, we propose a new method (DrugEx) to discover de novo drug-like molecules. DrugEx is an RNN model (generator) trained through a special exploration strategy integrated into reinforcement learning. As a case study we applied our method to design ligands against the adenosine A2A receptor. From ChEMBL data, a machine learning model (predictor) was created to predict whether generated molecules are active or not. Based on this predictor as the reward function, the generator was trained by reinforcement learning without any further data. We then compared the performance of our method with two previously published methods, REINVENT and ORGANIC. We found that candidate molecules our model designed that were predicted to be active, had a larger chemical diversity, and better covered the chemical space of known ligands compared to the state-of-the-art.

Poster presented at the RSC BMCS & AgriNet Agriscience Symposium: Biological and Chemical Approaches Towards Combating Resistance in Agriculture(http://www.rsc.org/ConferencesAndEvents/conference/alldetails.cfm?evid=113227)

Poster presented at the 10th ICCS (http://www.int-conf-chem-structures.org/)

Poster presented at the 2013 FIGON DMD (http://www.figondmd.nl/)

CyberKnife® stereotactic body radiation therapy is used to treat extracranial tumour sites that move with respiration. It has also been employed for the successful treatment of prostate cancer, using the image-tracking CyberKnife® system to compensate for intrafraction movements resulting from peristaltic motion and bladder filling. Large sporadic motions can be compensated for using an online target motion monitoring and cybernetic correction strategy. Radio-opaque gold markers can be implanted in the bladder during transurethral resection and used for online image-tracking during radiation to compensate for bladder filling and target movements. Transurethral bladder resection followed by chemoradiation and a stereotactic CyberKnife® radiotherapy boost seems a promising approach for the treatment of invasive bladder cancer in heavily pre-treated patients or patients eligible for preservation strategies. In this case study of a patient with a previously irradiated pelvis, CyberKnife® radiotherapy was feasible and well tolerated, with disease control and non-altered functional results two years after treatment completion. CyberKnife® irradiation may also be considered for the conservative treatment of locally advanced T2–T4a N0 M0 bladder cancer with incomplete or uncertain transurethral resection.

International audience; BACKGROUND AND PURPOSE: During the first decade of the 21st century a number of important European randomized studies were published. In order to help shape clinical practice based on best scientific evidence from the literature, the International Conference on 'Multidisciplinary Rectal Cancer Treatment: Looking for an European Consensus' (EURECA-CC2) was organized in Italy under the endorsement of European Society of Medical Oncology (ESMO), European Society of Surgical Oncology (ESSO), and European Society of Therapeutic Radiation Oncology (ESTRO). METHODS: Consensus was achieved using the Delphi method. The document was available to all Committee members as a web-based document customized for the consensus process. Eight chapters were identified: epidemiology, diagnostics, pathology, surgery, radiotherapy and chemotherapy, treatment toxicity and quality of life, follow-up, and research questions. Each chapter was subdivided by a topic, and a series of statements were developed. Each member commented and voted, sentence by sentence thrice. Sentences upon which an agreement was not reached after voting round # 2 were openly debated during a Consensus Conference in Perugia (Italy) from 11 December to 13 December 2008. A hand-held televoting system collected the opinions of both the Committee members and the audience after each debate. The Executive Committee scored percentage consensus based on three categories: "large consensus", "moderate consensus", and "minimum consensus". RESULTS: The total number of the voted sentences was 207. Of the 207, 86% achieved large consensus, 13% achieved moderate consensus, and only 3 (1%) resulted in minimum consensus. No statement was disagreed by more than 50% of the members. All chapters were voted on by at least 75% of the members, and the majority was voted on by >85%. CONCLUSIONS: This Consensus Conference represents an expertise opinion process that may help shape future programs, investigational protocols, and guidelines for staging and treatment of rectal cancer throughout Europe.