Related Experiment Video
Updated: Oct 3, 2025
03:14
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
705
Machine vs. Radiologist-Based Translations of RadLex: Implications for Multi-language Report Interoperability
Christian J Park1, Paul H Yi2, Hussain Al Yousif3
1Department of Radiology, Penn State Health Milton S. Hershey Medical Center, Hershey, PA, USA. cpark@pennstatehealth.psu.edu.
Journal of Digital Imaging
|February 15, 2022
Summary
Google Translate shows promise for translating radiology terms between English and German, especially when combined with human review. While English to German translation accuracy is high (95.7%), German to English translation needs further improvement.
Area of Science:
- Medical Informatics
- Natural Language Processing
- Radiology
Background:
- Accurate translation of medical terminology is crucial for international collaboration and data sharing in radiology.
- Existing machine translation tools require rigorous evaluation for specialized vocabularies like RadLex.
Purpose of the Study:
- To assess the feasibility and accuracy of Google Translate for English-to-German and German-to-English translation of the RadLex ontology.
- To compare machine translation concordance with official RadLex terms and evaluate clinical utility through human review.
Main Methods:
- Character-by-character concordance analysis (string matching) was used to compare Google Translate outputs with official RadLex terms.
- Radiologists reviewed non-concordant translations to assess clinical utility and understandability.
- Success rates were calculated by combining string matching and human review data.
Main Results:
- English to German translation showed 5.4% concordance via string matching, with a combined success rate of 95.7% after human review.
- German to English translation showed 6.9% concordance via string matching, with a combined success rate of 47.8% after human review.
- Human review revealed high understandability for English-to-German (95.5%) but lower for German-to-English (43.9%) non-concordant translations.
Conclusions:
- Google Translate, particularly when supplemented by human review, can be a valuable tool for translating radiology reports and supporting natural language processing tasks.
- String matching alone is insufficient for evaluating machine translation quality; human assessment is essential.
- Translation accuracy is direction-dependent, with English-to-German performing significantly better than German-to-English.
More Related Videos
Related Concept Videos
02:07Improving Translational Accuracy
2.8K
2.8K
02:28Leaky Scanning
5.3K
During most eukaryotic translation processes, the small 40S ribosome subunit scans an mRNA from its 5' end until it encounters the first start AUG codon. The large 60S ribosomal subunit then joins the smaller one to initiate protein synthesis. The location of the translation initiation is largely determined by the nucleotides near the start codon as there may be multiple translation initiation sites present on the mRNA. Marilyn Kozak discovered that the sequence RCCAUGG (where R...
5.3K
02:24Ribosome Profiling
3.7K
Ribosome profiling or ribo-sequencing is a deep sequencing technique that produces a snapshot of active translation in a cell. It selectively sequences the mRNAs protected by ribosomes to get an insight into a cell’s translation landscape at any given point in time.
Applications of ribosome profiling
Ribosome profiling has many applications, including in vivo monitoring of translation inside a particular organ or tissue type and quantifying new protein synthesis levels.
The technique...
Applications of ribosome profiling
Ribosome profiling has many applications, including in vivo monitoring of translation inside a particular organ or tissue type and quantifying new protein synthesis levels.
The technique...
3.7K