2 results
OP22 Benchmarking Of Population-Based Childhood Cancer Survival By Toronto Stage: Know The Differences To Propose Effective Interventions
- Rosalia Ragusa, Dott Fabio Didonè, Laura Botta, Antonina Torrisi, Maria Alessandra Bellia, Gemma Gatta, BENCHISTA Italy working group
-
- Journal:
- International Journal of Technology Assessment in Health Care / Volume 39 / Issue S1 / December 2023
- Published online by Cambridge University Press:
- 14 December 2023, p. S7
-
- Article
-
- You have access Access
- Export citation
-
Introduction
Pediatric cancers are rare tumors, heterogeneous in location and biologically very different from adult cancers. Documented survival variation across European countries and Italian regions shows that there is still room for further improvement by reducing inequalities. We aim to understand why there are differences in survival. The BENCHISTA-ITA project (National Benchmarking of Childhood Cancer Survival by Stage at diagnosis), that is the Italian twin project of the International BENCHISTA, collects stage at diagnosis of solid pediatric tumors, according to the Toronto Guidelines. We will compare how far the cancer has spread at diagnosis and test if differences in tumor stage explain any survival differences between Italian regions.
MethodsThe project study involved the stage distribution and the survival of 9 pediatric solid tumors diagnosed between 2013 and 2017 in Italy. All patients therefore had at least 3 years of follow-up in 2021 for life-stage definition. The study involves the identification of all new diagnoses of cancer, evaluation of the clinical documentation of cases eligible for research, and international classification and coding. Analyses of stage distribution and survival rates for each tumor type will be described.
ResultsData from 35 population-based cancer registries from 18 out of 20 Italian regions were collected covering about 84 percent of the Italian child population. In particular, data on: imaging/examination performed before any treatment; source used for staging; primary treatment defined as given within one year from diagnosis; relapse/ recurrence/ progression; follow up and status of life. The study tested the applicability of the Toronto Guidelines as a tool to obtain population-level comparable stage information for childhood cancers. There were 1,343 cases collected (242 Neuroblastoma, 124 Wilms Tumour, 145 Medulloblastoma, 148 Osteosarcoma, 135 Ewing sarcoma, 115 Rhabdomyososarcoma, 54 Ependymoma, 47 Retinoblastoma, 333 Astrocytoma). Toronto stage could be assigned in more than 90 percent in the majority of tumors. Tumors in which it was more difficult to assign the stage using the Toronto staging guidelines were ependymoma, astrocytoma, and retinoblastoma. It was easier to retrieve data for patients in the 0-14 years of age range than adolescents (14-18 years). Differences in stage distribution and survival differences between regional grouping were presented.
ConclusionsThe Italian BENCHISTA project, improving the connection between pediatric cancer registries, aims to improve care of children with cancer across the nation, reducing possible disparities.
The wide adoption of the Toronto Guidelines will facilitate international comparative incidence studies, strengthen the interpretation of survival data, and contribute to more appropriate solutions to improve childhood cancer outcomes.
OP135 Machine Learning And Cancer Registry: Evaluation Of The Effectiveness Of Case Coding
- Carmelo Ettore Viscosi, Alessia Anna Di Prima, Antonina Torrisi, Antonietta Alfia Torrisi, Margherita Ferrante, Rosalia Ragusa
-
- Journal:
- International Journal of Technology Assessment in Health Care / Volume 39 / Issue S1 / December 2023
- Published online by Cambridge University Press:
- 14 December 2023, p. S39
-
- Article
-
- You have access Access
- Export citation
-
Introduction
Machine learning (ML) algorithms are computational procedures that use pattern recognition and inference by learning from previously categorized documents to predict the category to which a new document belongs. The role of machine learning within cancer registries remains unclear given the lack of in-depth testing and guidance from health technology assessment (HTA) agencies. We evaluated the effectiveness of coding new cases through machine learning at the Integrated Cancer Registry.
MethodsThe Integrated Cancer Registry covers the eastern area of Sicily in Italy, which has an annual average incidence of about 10,000 cases of malignant neoplasm. Potential new cancer cases were retrieved from pathology services and processed by pathologists who confirmed the neoplastic nature of supposed cases and specified the morphological type and location of the tumors. The current method involves identification by reading the free-text report when International Classification Diseases for Oncology information was not provided. We used the new Microsoft ML.Net Library, a framework developed in response to the challenge of facilitating machine learning pipeline utilization in large software applications. A total of 1,050,952 free-text pathology reports published from 2003 to 2018 were selected separately from all Sicilian pathology services and uploaded to machine learning software that explored eight binary classification algorithms.
ResultsWe evaluated each algorithm’s performance by calculating metrics (the number of true positives, true negatives, false positives, and false negatives) from the classification procedure applied to the test dataset. The metrics used were accuracy, F1 score, and area under the receiver operating characteristic curve. With a test set of around 210,000 text diagnoses, each algorithm reached an F1 score of up to 95 percent.
ConclusionsMachine learning algorithms capture relevant information about tumors from free-text pathology reports, optimizing the process and reducing waste. With the help of machine learning systems, cancer registries can provide more timely data for research and evaluation of all types of new cancer technologies (drugs, devices, radiology and radiotherapy equipment, diagnostic devices, robotic surgery, and vaccines).