Publications
106 results found
Hart MG, Hunter A, Hawkins N, et al., 2018, First-line treatments for people with single or multiple brain metastases, Cochrane Database of Systematic Reviews, Vol: 2018
© 2018 The Cochrane Collaboration. This is a protocol for a Cochrane Review (Intervention). The objectives are as follows: To compare the safety and efficacy of surgery, radiotherapy, and chemotherapy as first-line treatment for people with single or multiple brain metastases, either alone or in combination.
Brown NF, Williams M, Arkenau H-T, et al., 2018, A study of the focal adhesion kinase inhibitor GSK2256098 in patients with recurrent glioblastoma with evaluation of tumor penetration of [C-11]GSK2256098, NEURO-ONCOLOGY, Vol: 20, Pages: 1634-1642, ISSN: 1522-8517
Rabinowicz S, Butz R, Hommerson A, et al., 2018, CSBN: a hybrid approach for survival time prediction with missing data, AALTD: 3nd ECML/PKDD Workshop on Advanced Analytics and Learning on Temporal Data
CSBN: A Hybrid Approach For Survival TimePrediction With Missing DataSimon Rabinowicz1, Raphaela Butz2,3, Arjen Hommersom3,4, and Matt Williams5,61Faculty Of Medicine, Imperial College London, UK2Institute for Computer Science, TH K ̈oln, Germany3Department of Computer Science, Open University of the Netherlands4Department of Software Science, Radboud University, The Netherlands5Department of Radiotherapy, Charing Cross Hospital, London, UK6Computational Oncology Laboratory, Imperial College London, UKAbstract.Survival prediction models most commonly use Cox Proportional Hazards (CPH) models, and are frequently used in medical statistics and clinical practice. However, such models underperform when the predictor variables are missing. By building Bayesian networks we automatically construct a model with the most important risk factors and relationships between risk factors and Bayesian networks are able to infer the likely values of missing data. We therefore propose a hybrid solution, consisting of a CPH model and a BN, where the predictive variables in the CPH model are the child nodes of a BN, which we call CSBN. We learn the CPH and BN models separately, using standard techniques, with the only constraint being that the variables that are predictors in the CPH model are child nodes in the BN. This allows us to fuse the two models, using the predictors of the CPH models as the join points. We test our approach by examining the performance of the CPH model, against the hybrid CSBN model, using both complete data cases and in cases with missing data. We calculate the performance of the survival prediction for both CPH and CSBN using the C-index and a normalised error function as metrics. For the CPH model, predictive error was significantly larger for missing data (±3120.8 days) compared to complete data (±1171.5 days;p= 3.6e−07). This was also true for the CSBN±1387.3 days for missing data compared with±1171.5 days with com
Wong S-L, Ricketts K, Royle G, et al., 2018, A methodology to extract outcomes from routine healthcare data for patients with locally advanced non-small cell lung cancer, BMC Health Services Research, Vol: 18, ISSN: 1472-6963
BACKGROUND: Outcomes for patients in UK with locally advanced non-small cell lung cancer (LA NSCLC) are amongst the worst in Europe. Assessing outcomes is important for analysing the effectiveness of current practice. However, data quality is inconsistent and regular large scale analysis is challenging. This project investigates the use of routine healthcare datasets to determine progression free survival (PFS) and overall survival (OS) of patients treated with primary radical radiotherapy for LA NSCLC. METHODS: All LA NSCLC patients treated with primary radical radiotherapy in a 2 year period were identified and paired manual and routine data generated for an initial pilot study. Manual data was extracted information from hospital records and considered the gold standard. Key time points were date of diagnosis, recurrence, death or last clinical encounter. Routine data was collected from various data sources including, Hospital Episode Statistics, Personal Demographic Service, chemotherapy data, and radiotherapy datasets. Relevant event dates were defined by proxy time points and refined using backdating and time interval optimization. Dataset correlations were then tested on key clinical outcome indicators to establish if routine data could be used as a reliable proxy measure for manual data. RESULTS: Forty-three patients were identified for the pilot study. The manual data showed a median age of 67 years (range 46- 89 years) and all patients had stage IIIA/B disease. Using the manual data, the median PFS was 10.78 months (range 1.58-37.49 months) and median OS was 16.36 months (range 2.69-37.49 months). Based on routine data, using proxy measures, the estimated median PFS was 10.68 months (range 1.61-31.93 months) and estimated median OS was 15.38 months (range 2.14-33.71 months). Overall, the routine data underestimated the PFS and OS of the manual data but there was good correlation with a Pearson correlati
Williams M, Morton CE, 2018, Computational Medicine: Coding for Medics, Publisher: Elsevier, ISBN: 9780702076039
Majewska P, Ioannidis S, Raza MH, et al., 2017, Postprogression survival in patients with glioblastoma treated with concurrent chemoradiotherapy: a routine care cohort study, CNS Oncology, Vol: 6, Pages: 307-313, ISSN: 2045-0907
Glioblastoma is the commonest malignant brain tumor in adults. Most patients develop progressive disease before they die. However, survival after developing progressive disease is infrequently reported. We identified patients with histologically proven disease who were treated with concurrent chemoradiotherapy during 2006–2013. We analyzed overall survival (OS), progression-free survival and postprogression survival (PPS) in relation to age, O6-methylguanine-DNA methyltransferase promoter methylation and extent of surgical resection. We identified 166 patients. Median survival was 13.5 months; 2-year OS was 21.7%. Median progression-free survival and PPS were 7.03 and 4.53 months, respectively. Age and extent of surgical resection were correlated with OS. Only the extent of surgical resection was associated with PPS. Our work suggests that the established prognostic factors for glioblastoma do not appear to help predict PPS.
Kelly C, Majewska P, Ioannidis S, et al., 2017, Estimating progression-free survival in patients with glioblastoma using routinely collected data, Journal of Neuro-Oncology, Vol: 135, Pages: 621-627, ISSN: 0167-594X
Glioblastoma (GBM) represents 80% of all primarymalignant brain tumours in adults. Prognosis is poor,and there is a clear correlation between disease progressionand deterioration in functional status. In this pilot study weassess whether we can estimate disease progression andprogression free survival (PFS) from routinely collectedelectronic healthcare data. We identified fifty patients withglioblastoma who had chemo-radiotherapy. For each patientwe manually collected a reference data set recording demographics,surgery, radiotherapy, chemotherapy, follow-up anddeath. We also obtained an electronic routine data set for eachpatient by combining local data on chemotherapy/radiotherapyand hospital admissions. We calculated overall survival(OS) and PFS using the reference data set, and estimatedthem using the routine data sets using two different methods,and compared the estimated measures with the referencemeasures. Overall survival was 68% at 1 year and medianOS was 12.8 months. The routine data correctly identifiedprogressive disease in 37 of 40 patients and stable disease in 7 of 10 patients. PFS was 7.4 months and the estimated PFSusing routine data was 9.1 and 7.8 months with methods 1and 2 respectively. There was acceptable agreement betweenreference and routine data in 49 of 50 patients for OS and 35of 50 patients for PFS. The event of progression, subsequenttreatment and OS are well estimated using our approach, butPFS estimation is less accurate. Our approach could refineour understanding of the disease course and allow us to reportPFS, OS and treatment nationally.
Reardon CH, Zienius K, Wood S, et al., 2017, Ketogenic diet for primary brain and spinal cord tumours, Cochrane Database of Systematic Reviews, Vol: 2017, ISSN: 1465-1858
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows: To evaluate the effectiveness and tolerability of dietary intervention with the ketogenic diet (KD) for the management of brain and spinal cord tumours, including the effect on brain-tumour related epilepsy.
Williams M, Hommersom A, Butz R, et al., 2017, A Prognostic Model of Glioblastoma Multiforme Using Survival Bayesian Networks, Conference on Artificial Intelligence in Medicine in Europe, Publisher: Springer
Bayesian networks are attractive for developing prognostic models in medicine, due to the possibility for modelling the multivariate relationships between variables that come into play in the care process. In practice, the development of these models is hindered due to the fact that medical data is often censored, in particular the survival time. In this paper, we propose to directly integrate Cox proportional hazards models as part of a Bayesian network. Furthermore, we show how such Bayesian network models can be learned from data, after which these models can be used for probabilistic reasoning about survival. Finally, this method is applied to develop a prognostic model for Glioblastoma Multiforme, a common malignant brain tumour.
Williams M, Treasure P, Greenberg D, et al., 2016, Surgeon volume and 30 day mortality for brain tumours in England, British Journal of Cancer, Vol: 115, Pages: 1379-1382, ISSN: 1532-1827
background: There is evidence that surgeons who perform more operations have better outcomes. However, in patients with brain tumours, all of the evidence comes from the USA.methods: We examined all English patients with an intracranial neoplasm who had an intracranial resection in 2008–2010. We included surgeons who performed at least six operations over 3 years, and at least one operation in the first and last 6 months of the period.results: The analysis data set comprised 9194 operations, 163 consultant neurosurgeons and 30 centres. Individual surgeon volumes varied widely (7–272; median=46). 72% of operations were on the brain, and 30 day mortality was 3%. A doubling of surgeon load was associated with a 20% relative reduction in mortality. Thirty day mortality varied between centres (0·95–8·62%) but was not related to centre workload.conclusions: Individual surgeon volumes correlated with patient 30 day mortality. Centres and surgeons in England are busier than surgeons and centres in the USA. There is no relationship between centre volume and 30 day mortality in England. Services in the UK appear to be adequately arranged at a centre level, but would benefit from further surgeon sub-specialisation.
Mocanu A, Fan X, Toni F, et al., 2016, Online argumentation-based platform for recommending medical literature, 4th International Workshop, CIMA 2014, Publisher: Springer, Pages: 97-115, ISSN: 2190-3018
In medical practice, choosing the correct treatment is a key problem [1]. In this work, we present an online medical recommendation system, RecoMedic, that selects most relevant medical literature for patients. RecoMedic maintains a medical literature repository in which users can add new articles, query existing articles, compare articles and search articles guided by patient information. RecoMedic uses argumentation to accomplish the article selection. Thus, upon identifying relevant articles, RecoMedic also explains its selection. RecoMedic can be deployed using single-agent as well as multi-agent implementations. The developed system has been experimented with by senior medical Ph.D students from SouthernMedical University in China.
Williams M, Hunter A, 2016, Aggregation of Clinical Evidence Using Argumentation: A Tutorial Introduction, Foundations of Biomedical Knowledge Representation Methods and Applications, Editors: Hommersom, Lucas, Publisher: Springer, Pages: 317-337, ISBN: 9783319280073
We present a novel argumentation-based system for reasoning with the summarised results of clinical trials, and explore its use in a lung cancer context.
Williams M, Liu ZW, Hunter A, et al., 2015, An updated systematic review of lung chemo-radiotherapy using a new evidence aggregation method, LUNG CANCER, Vol: 87, Pages: 290-295, ISSN: 0169-5002
- Author Web Link
- Cite
- Citations: 6
Brodbelt A, Greenberg D, Winters T, et al., 2015, Glioblastoma in England: 2007-2011, EUROPEAN JOURNAL OF CANCER, Vol: 51, Pages: 533-542, ISSN: 0959-8049
- Author Web Link
- Open Access Link
- Cite
- Citations: 126
Ricketts K, Williams M, Liu Z-W, et al., 2014, Automated estimation of disease recurrence in head and neck cancer using routine healthcare data, COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, Vol: 117, Pages: 412-424, ISSN: 0169-2607
- Author Web Link
- Open Access Link
- Cite
- Citations: 11
Williams M, Singer RA, Lerner A, 2014, A simple technique to estimate best- and worst-case survival in patients with metastatic colorectal cancer treated with chemotherapy, ANNALS OF ONCOLOGY, Vol: 25, Pages: 2014-2019, ISSN: 0923-7534
- Author Web Link
- Open Access Link
- Cite
- Citations: 2
Savage P, Sharkey R, Kua T, et al., 2014, Malignant spinal cord compression: NICE guidance, improvements and challenges., QJM : monthly journal of the Association of Physicians, Vol: 107, Pages: 277-82, ISSN: 1460-2393
BACKGROUND AND AIM: Malignant spinal cord compression (mSCC) is one of the most serious complications of cancer. Recent NICE guidance has aimed to improve patient pathways and outcomes for patients with mSCC. We have examined the current presentations, management and outcomes for patients with mSCC in West London following the implementation of the NICE guidance. MATERIALS AND METHODS: The electronic records and clinical notes were reviewed for all patients assessed for confirmed or potential mSCC at Charing Cross Hospital in 2012. Details on the number of referrals, the proportion with confirmed mSCC, the cancer diagnosis, treatment and outcome were analysed. RESULTS: 191 patients were reviewed with 127 (66%) cases of confirmed mSCC. The commonest tumour types were prostate cancer (26 cases), lung cancer (26), breast cancer (21) and kidney cancer (15). 21% of the patients had no previous cancer diagnosis; mSCC was their presenting diagnostic event. Radiotherapy was the predominant management, 24% of the patients had first line surgical treatment. At presentation 62% of patients were either chair or bed bound. Treatment brought important mobility benefits to all patients groups with 20% of the initially chair or bed bound patients leaving the hospital with independent mobility. CONCLUSION: Enhanced patients pathways with ease of access, rapid assessment and prompt treatment can improve outcomes. Despite these pathways many patients still present with gross motor impairment and over 20% have no previous diagnosis of cancer. Ongoing work to maintain awareness for patients and primary care of the diagnosis and emergency pathways is essential to optimize outcomes.
Craven R, Toni F, Williams M, 2014, Graph-based dispute derivations in assumption-based argumentation, Pages: 46-62, ISSN: 0302-9743
Arguments in structured argumentation are usually defined as trees. This introduces both conceptual redundancy and inefficiency in standard methods of implementation. We introduce rule-minimal arguments and argument graphs to solve these problems, studying their use in assumption-based argumentation (ABA), a well-known form of structured argumentation. In particular, we define a new notion of graph-based dispute derivations for determining acceptability of claims under the grounded semantics in ABA, study formal properties and present an experimental evaluation thereof. © 2014 Springer-Verlag Berlin Heidelberg.
Craven R, Toni F, Williams M, 2014, Graph-based dispute derivations in assumption-based argumentation, Second International Workshop, TAFA 2013, Publisher: Springer, Pages: 46-62, ISSN: 0302-9743
Arguments in structured argumentation are usually defined as trees. This introduces both conceptual redundancy and inefficiency in standard methods of implementation. We introduce rule-minimal arguments and argument graphs to solve these problems, studying their use in assumption-based argumentation (ABA), a well-known form of structured argumentation. In particular, we define a new notion of graph-based dispute derivations for determining acceptability of claims under the grounded semantics in ABA, study formal properties and present an experimental evaluation thereof. © 2014 Springer-Verlag Berlin Heidelberg.
Fan X, Toni F, Mocanu A, et al., 2014, Dialogical Two-Agent Decision Making with Assumption-based Argumentation, International Conference on Autonomous Agents and Multiagent Systems (AAMAS), Publisher: ASSOC COMPUTING MACHINERY, Pages: 533-540
- Author Web Link
- Cite
- Citations: 21
Woolf DK, Williams M, Goh CL, et al., 2013, Fractionated stereotactic radiotherapy for acoustic neuromas: long-term outcomes., Clinical oncology (Royal College of Radiologists (Great Britain)), Vol: 25, Pages: 734-8, ISSN: 1433-2981
AIMS: Acoustic neuromas are rare, benign intracranial tumours. There are a variety of treatment options, with no clear optimal management strategy and wide variation in treated outcomes. We report the outcomes from a 15 year cohort of patients treated at our centre using fractionated stereotactic radiotherapy (52.5 Gy in 25 fractions). MATERIALS AND METHODS: We analysed a retrospective case series. Patients were identified from patient records and a retrospective review of case notes and imaging reports was undertaken. We assessed tumour response using RECIST criteria and recorded toxicity. Progression-free survival was estimated using the Kaplan-Meier method. The study was conducted according to the STROBE guidelines. RESULTS: In total, 93 patients were identified; 83 patients had follow-up data, with a median follow-up period of 5.7 years. The overall control rate using RECIST criteria was 92%. Data on complications were available for 90 patients, with six (7%) experiencing a reduction in hearing, one (1%) developing trigeminal nerve dysfunction and one (1%) a deterioration in facial nerve function. Other toxicities included four (4%) patients who developed hydrocephalus, requiring the placement of a shunt and one (1%) patient who developed radiation brainstem necrosis. After further evaluation this patient was deemed to have been treated within acceptable dose constraints. CONCLUSION: These data suggest that a good control rate of acoustic neuromas is achievable using fractionated stereotactic radiotherapy to a dose of 52.5 Gy in 25 fractions. Toxicity is considered acceptable but the episode of radiation brainstem necrosis remains of concern and is the subject of further work.
Williams M, Woolf D, Dickson J, et al., 2013, Routine clinical data predict survival after palliative radiotherapy: an opportunity to improve end of life care., Clinical oncology (Royal College of Radiologists (Great Britain)), Vol: 25, Pages: 668-73, ISSN: 1433-2981
AIMS: Estimating the prognosis of cancer patients with incurable disease remains an important and difficult task for clinicians. Radiotherapy is a commonly used modality for palliation of symptoms, and we investigated whether we could predict differences in overall survival after the first course of palliative radiotherapy using routinely available data. MATERIALS AND METHODS: We examined variations in survival in 1226 patients after their first course of palliative radiotherapy in relation to cancer type, site treated, age, gender and socioeconomic status, and developed a multivariate model based on these. RESULTS: The median overall survival after the first course of palliative radiotherapy was 5.2 months. Large differences in survival were seen, depending on the primary tumour and the site treated. Survival was much better in those with breast (median overall survival 11.4 months) or prostate cancer (8.4 months, hazard ratio = 1.3) than in those with oesophageal/gastro-oesophageal junctional tumours (4.6 months, hazard ratio = 2.3) or lung (3.9 months, hazard ratio = 2.5). The treated site was an important prognostic factor (primary tumour versus bone metastases, hazard ratio = 1.3; versus brain metastases, hazard ratio = 2.1). CONCLUSIONS: The median overall survival after a first course of palliative radiotherapy was less than 6 months. Simple data, provided as part of routine radiotherapy practice, clearly discriminate between patients with very different prognoses. Such data could therefore be used to trigger appropriate end of life care.
Liu Z-W, Fitzke H, Williams M, 2013, Using routine data to estimate survival and recurrence in head and neck cancer: our preliminary experience in twenty patients., Clinical otolaryngology : official journal of ENT-UK ; official journal of Netherlands Society for Oto-Rhino-Laryngology & Cervico-Facial Surgery, Vol: 38, Pages: 334-9, ISSN: 1749-4486
Williams M, Swampillai A, Osborne M, et al., 2013, Squamous cell carcinoma antigen A Potentially Useful Prognostic Marker in Squamous Cell Carcinoma of the Anal Canal and Margin, CANCER, Vol: 119, Pages: 2391-2398, ISSN: 0008-543X
- Author Web Link
- Cite
- Citations: 25
Kuku S, Williams M, McCormack M, 2013, Adjuvant therapy in stage III endometrial cancer: treatment outcomes and survival. a single-institution retrospective study., International journal of gynecological cancer : official journal of the International Gynecological Cancer Society, Vol: 23, Pages: 1056-64, ISSN: 1525-1438
OBJECTIVE: As adjuvant treatment of advanced-stage endometrial cancer remains undefined, we sought to review and describe the outcomes of patients with International Federation of Obstetrics and Gynecology stage III endometrial cancer treated with chemotherapy and/or radiotherapy after primary surgery. METHODS: We conducted a retrospective cohort study of patients with stage III disease treated at University College London Hospitals from 2002 to 2009. Patients were eligible if they received adjuvant treatment at our center. We excluded those with any synchronous gynecologic tumor and patients who underwent surgery but not adjuvant treatment at the center. RESULTS: Stages IIIA, IIIB, and IIIC tumors accounted for 60%, 10%, and 30%, respectively. The median age was 67 years (range, 37-94 years). Sixty-five percent were pure endometrioid tumors, and 65% were high-grade (grade 3) tumors. Eighty-one patients received adjuvant treatment, 9% received chemotherapy alone, 28% received radiotherapy alone, and 63% received sequential combined chemotherapy followed by external beam radiotherapy with vaginal vault brachytherapy. In multivariate analysis, there was a significant difference between the adjuvant treatment groups for disease-free survival (DFS) and overall survival (OS) with those who received chemotherapy (DFS: P = 0.0001; hazard ratio [HR], 6.2; 95% confidence interval [CI], 2.47-15.8; OS: P = 0.003; HR, 6.0; CI, 2.2-16.6) or radiotherapy alone (DFS: P = 0.06; HR, 1.88; CI, 0.97-3.7; OS: P = 0.025; HR, 2.1; CI, 1.1-4.1) having a poorer survival compared to combined treatment. Overall survival at 3 years and 5 years were 57% and 47%, respectively, for all 81 patients who received any adjuvant treatment. CONCLUSIONS: Sequential combined adjuvant chemotherapy and radiotherapy may be associated with a significant improvement in survival compared with chemotherapy or radiotherapy alone. Univariate and multivariate analysis showed that advanced age, high grade, and pres
, 2013, Computational Logic in Multi-Agent Systems, Departmental Report, Publisher: Springer Berlin Heidelberg
Fan X, Craven R, Singer R, et al., 2013, Assumption-Based Argumentation for Decision-Making with Preferences: A Medical Case Study, Publisher: SPRINGER-VERLAG BERLIN
- Author Web Link
- Cite
- Citations: 14
Hunter A, Williams M, 2012, Aggregating evidence about the positive and negative effects of treatments., Artificial intelligence in medicine, Vol: 56, Pages: 173-90, ISSN: 1873-2860
OBJECTIVES: Evidence-based decision making is becoming increasingly important in healthcare. Much valuable evidence is in the form of the results from clinical trials that compare the relative merits of treatments. In this paper, we present a new framework for representing and synthesizing knowledge from clinical trials involving multiple outcome indicators. METHOD: The framework generates and evaluates arguments for claiming that one treatment is superior, or equivalent, to another based on the available evidence. Evidence comes from randomized clinical trials, systematic reviews, meta-analyses, network analyses, etc. Preference criteria over arguments are used that are based on the outcome indicators, and the magnitude of those outcome indicators, in the evidence. Meta-arguments attacks arguments that are based on weaker evidence. RESULTS: We evaluated the framework with respect to the aggregation of evidence undertaken in three published clinical guidelines that involve 56 items of evidence and 16 treatments. For each of the three guidelines, the treatment we identified as being superior using our method is a recommended treatment in the corresponding guideline. CONCLUSIONS: The framework offers a formal approach to aggregating clinical evidence, taking into account subjective criteria such as preferences over outcome indicators. In the evaluation, the aggregations obtained showed a good correspondence with published clinical guidelines. Furthermore, preliminary computational studies indicate that the approach is viable for the size of evidence tables normally encountered in practice.
Williams M, Liu ZW, Woolf D, et al., 2012, Change in platelet levels during radiotherapy with concurrent and adjuvant temozolomide for the treatment of glioblastoma: a novel prognostic factor for survival., Journal of cancer research and clinical oncology, Vol: 138, Pages: 1683-8, ISSN: 1432-1335
BACKGROUND: Radiotherapy plus concomitant and adjuvant temozolomide (RCAT) is now standard treatment for grade IV glioblastoma (GBM). We report the results from our 7 years experience of using RCAT, and the potential role of a change in platelet count as a prognostic factor. METHODS: We identified all patients with biopsy-proven GBM who received RCAT at the Royal Free Hospital between 2002 and 2009. We extracted data on demographic, tumour and treatment variables and overall survival and conducted univariate analyses on the association of the baseline factors with survival, and included those that were significant in a multivariate model. We then conducted exploratory analyses on the impact of changes in haematological parameters and overall survival. RESULTS: A total of eighty-four patients were included in the final analysis. Median overall survival in our study was 17.6 months. Overall survival rate at 1 year and 2 years were 70 and 36 %, respectively. Platelet counts were seen to fall when measured from baseline to beginning of week 6. A decrease in platelet count from baseline to week 6 was associated with longer survival (p = 0.006), and this remains significant when adjusted for known prognostic factors. CONCLUSION: Our study shows the survival benefit seen in the phase III trial is reproducible in clinical practice. In addition, decreased platelet count during concurrent radiotherapy and temozolomide appears to correlate with prolonged survival, a finding that warrants further investigation.
Williams M, Hunter A, 2012, CAFE: A computational Framework for Evidence
We present an implementation of a framework for reasoning with the summarised results of clinical trials.The software is a partial implementation of the work described in a series of papers by Williams & Hunter (and others). The most concise description is given in the 2012 AI in Medicine paper (Aggregating evidence about the positive and negative effects of treatments, Artificial Intelligence in Medicine 56:173-190)The software is available as open-source software, hosted by Google code. For more information, please contact Anthony Hunter or Matt Williams.
This data is extracted from the Web of Science and reproduced under a licence from Thomson Reuters. You may not copy or re-distribute this data in whole or in part without the written consent of the Science business of Thomson Reuters.