Introduction
Dear Editor, we would like to discuss `Large Language
Models as a Rapid and Objective Tool for Pathology Report
Data Extraction[
1]`. According to Bolat et al., it is
possible that the large massive language models might be
a new useful thing in pathology. Although employing new
language models to extract data from pathology reports is
a promising strategy, there are a few flaws in the process
that need to be fixed. The possibility of errors during the
data extraction procedure is one of the main worries. Errors
in the retrieved data may result from large language
models, such as ChatGPT and Google Bard, misinterpreting
subtle medical terms or nuances in pathology
reports. Inaccurate data may seriously affect the results
of research and jeopardize the validity of study findings.
Additionally, as these algorithms are processing sensitive
patient data, using AI for data extraction in pathology
reports may give rise to privacy and security problems.
Potential avenues for future research in this area might include
enhancing the precision and dependability of AI-assisted
data extraction in pathology reports. Addressing the
shortcomings of general-purpose models such as ChatGPT
and Google Bard may be possible through the development
of specialized language models trained only for the medical
sector. Ensuring the quality of the data can also be achieved
by manually reviewing the retrieved data by pathologists to
validate its accuracy. Additionally, investigating the application
of natural language processing methods to decipher and
comprehend pathology reports` intricate terminology could
improve the efficiency of AI algorithms for data extraction.
The potential impact of AI-assisted data extraction on
clinical practice is one area that is frequently disregarded
in the literature. Although academic research is usually the
main focus, clinical decision-making may be significantly
impacted by the use of AI for pathology report processing.
Artificial Intelligence (AI) has the potential to enhance
patient outcomes by facilitating healthcare providers` decision-
making process by automating the extraction and
analysis of data from pathology reports. The utilization of pathology reports in patient care might be completely
transformed by incorporating AI technology into clinical
workflows, which would ultimately result in more individualized
and efficient treatment plans.
Overall, even though using AI-assisted data extraction in
pathology reports has the potential to improve research
speed and precision, it`s critical to address the drawbacks
and difficulties that come with this methodology. Future
studies should concentrate on enhancing the precision and
dependability of AI algorithms, verifying the data that has
been retrieved, and investigating the therapeutic applications
of AI in the interpretation of pathology reports. AI
technologies have the power to revolutionize pathology
and enhance patient outcomes by tackling these issues.
Since ChatGPT relies solely on user input from humans,
programming that deals with human behavior must be
carefully chosen[2,3].
Conflict of Interest
The authors declare that they have no conflict of interest.
Authorship Contributions
Concept: HP, VW, Design: HP, VW, Supervision: HP, VW, Materials:
HP, VW, Data collection and/or processing: HP, VW, Analysis and/or
interpretation: HP, VW, Literature search: HP, VW, Writing: HP, VW,
Approval: HP, VW.
