| 1. |
- Catanuto, Giuseppe, et al.
(author)
-
Natural Language Processing to Extract Meaningful Information from a Corpus of Written Knowledge in Breast Cancer : Transforming Books into Data
- 2023
-
In: Breast Care. - : S. Karger AG. - 1661-3791 .- 1661-3805. ; 18:3, s. 209-212
-
Journal article (peer-reviewed)abstract
- Introduction: Books and papers are the most relevant source of theoretical knowledge for medical education. New technologies of artificial intelligence can be designed to assist in selected educational tasks, such as reading a corpus made up of multiple documents and extracting relevant information in a quantitative way.Methods: Thirty experts were selected transparently using an online public call on the website of the sponsor organization and on its social media. Six books edited or co-edited by members of this panel containing a general knowledge of breast cancer or specific surgical knowledge have been acquired. This collection was used by a team of computer scientists to train an artificial neural network based on a technique called Word2Vec.Results: The corpus of six books contained about 2.2 billion words for 300d vectors. A few tests were performed. We evaluated cosine similarity between different words.Discussion: This work represents an initial attempt to derive formal information from textual corpus. It can be used to perform an augmented reading of the corpus of knowledge available in books and papers as part of a discipline. This can generate new hypothesis and provide an actual estimate of their association within the expert opinions. Word embedding can also be a good tool when used in accruing narrative information from clinical notes, reports, etc., and produce prediction about outcomes. More work is expected in this promising field to generate "real-world evidence." (c) 2023 S. Karger AG, Basel
|
|
| 2. |
- Kastora, Stavroula Lila, et al.
(author)
-
Outcomes of Different Quality of Life Assessment Modalities After Breast Cancer Therapy : A Network Meta-analysis
- 2023
-
In: JAMA Network Open. - : American Medical Association (AMA). - 2574-3805. ; 6:6
-
Research review (peer-reviewed)abstract
- IMPORTANCE: Improvement in clinical understanding of the priorities of patients with breast cancer (BC) regarding postoperative aesthetic outcomes (AOs) is needed.OBJECTIVE: To assess expert panel and computerized evaluation modalities against patient-reported outcome measures (PROMs), the gold standard of AO assessment, in patients after surgical management of BC.DATA SOURCES: Embase, MEDLINE, PsycINFO, PubMed, the Cochrane Central Register of Controlled Trials, the World Health Organization International Clinical Trials Registry Platform, and ClinicalTrials.gov were interrogated from inception through August 5, 2022. Search terms included breast conserving AND aesthetic outcome AND breast cancer. Ten observational studies were eligible for inclusion, with the earliest date of database collection on December 15, 2022.STUDY SELECTION: Studies with at least 1 pairwise comparison (PROM vs expert panel or PROM vs computerized evaluation with Breast Cancer Conservation Treatment cosmetic results [BCCT.core] software) were considered eligible if they included patients who received BC treatment with curative intent. Studies reporting solely on risk reduction or benign surgical procedures were excluded to ensure transitivity.DATA EXTRACTION AND SYNTHESIS: Two independent reviewers extracted study data with an independent cross-check from a third reviewer. The quality of included observational studies was assessed using the Newcastle-Ottawa Scale, and the level of evidence quality was assessed using the Grading of Recommendations Assessment, Development and Evaluation tool. Confidence in network meta-analysis results was analyzed with the Confidence in Network Meta-analysis semiautomated tool. Effect size was reported using random-effects odds ratios (ORs) and cumulative ratios of ORs with 95% credibility intervals (CrIs).MAIN OUTCOMES AND MEASURES: The primary outcome of this network meta-analysis was modality (expert panel or computer software) discordance from PROMs. Four-point Likert responses across PROMs, expert panel assessment, and BCCT.core evaluation of AOs were assessed.RESULTS: A total of 10 observational studies including 3083 patients (median [IQR] age, 59 [50-60] years; median [range] follow-up, 39.0 [22.5-80.5] months) with reported AOs were assessed and homogenized in 4 distinct Likert response groups (excellent, very good, satisfactory, and bad). Overall network incoherence was low (χ22 = 0.35; P = .83). Overall, panel and software modalities graded AO outcomes worse than PROMs. Specifically, for excellent vs all other responses, the panel to PROM ratio of ORs was 0.30 (95% CrI, 0.17-0.53; I2 = 86%) and the BCCT.core to PROM ratio of ORs was 0.28 (95% CrI, 0.13-0.59; I2 = 95%), while the BCCT.core to panel ratio of ORs was 0.93 (95% CrI, 0.46-1.88; I2 = 88%).CONCLUSIONS AND RELEVANCE: In this study, patients scored AOs higher than both expert panels and computer software. Standardization and supplementation of expert panel and software AO tools with racially, ethnically, and culturally inclusive PROMs is needed to improve clinical evaluation of the journey of patients with BC and to prioritize components of therapeutic outcomes.
|
|
