Evaluation of lexicon- and syntax-based negation detection algorithms using clinical text data

Manimaran, J.; Velmurugan, T.

doi:10.1515/bams-2017-0016

Cited by 3 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each grammatical rule is converted into a graph, and then matched against the typed dependency graph using a subgraph matching algorithm. The subgraph matching algorithm is

where m is the length of the input, and k is the vertex degree, and must be repeated for every rule [ 11 ].…”

Section: Resultsmentioning

confidence: 99%

“…One such approach, DEEPEN [ 10 ], operates upon concepts that NegEx determines to be negated. Other dependency-based algorithms make no use of NegEx, such as NegBio, negation-detection, and DepNeg [ [11] , [12] , [13] ], reported to exhibit an improved precision over syntactic approaches. However, an independent assessment showed that ConText maintained its performance over a novel dataset, while the other approaches did not (Goryachev).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

Slater

Bradlow

Motti

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

Negation detection is an important task in biomedical text mining. Particularly in clinical settings, it is of critical importance to determine whether findings mentioned in text are present or absent. Rule-based negation detection algorithms are a common approach to the task, and more recent investigations have resulted in the development of rule-based systems utilising the rich grammatical information afforded by typed dependency graphs. However, interacting with these complex representations inevitably necessitates complex rules, which are time-consuming to develop and do not generalise well. We hypothesise that a heuristic approach to determining negation via dependency graphs could offer a powerful alternative. We describe and implement an algorithm for negation detection based on grammatical distance from a negatory construct in a typed dependency graph. To evaluate the algorithm, we develop two testing corpora comprised of sentences of clinical text extracted from the MIMIC-III database and documents related to hypertrophic cardiomyopathy patients routinely collected at University Hospitals Birmingham NHS trust. Gold-standard validation datasets were built by a combination of human annotation and examination of algorithm error. Finally, we compare the performance of our approach with four other rule-based algorithms on both gold-standard corpora. The presented algorithm exhibits the best performance by f-measure over the MIMIC-III dataset, and a similar performance to the syntactic negation detection systems over the HCM dataset. It is also the fastest of the dependency-based negation systems explored in this study. Our results show that while a single heuristic approach to dependency-based negation detection is ignorant to certain advanced cases, it nevertheless forms a powerful and stable method, requiring minimal training and adaptation between datasets. As such, it could present a drop-in replacement or augmentation for many-rule negation approaches in clinical text-mining pipelines, particularly for cases where adaptation and rule development is not required or possible.

show abstract

“…Each grammatical rule is converted into a graph, and then matched against the typed dependency graph using a subgraph matching algorithm. The subgraph matching algorithm is

where m is the length of the input, and k is the vertex degree, and must be repeated for every rule [ 11 ].…”

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

Slater

Bradlow

Motti

et al. 2021

Computers in Biology and Medicine

View full text Add to dashboard Cite

show abstract

“…This work did not use the instructions given by the journal. Experimental work was published in the Indian Journal of Science and Technology by Manimaran and Velmurugan 17 . The researchers perfectly used the instructions discussed above in this article.…”

Section: Requirements Of Scientific Articlesmentioning

confidence: 99%

Reviewer Perspective on the Essentials of Scientific Experiments Based Research Articles

Velmurugan¹

2023

Trends Schol. Pub

View full text Add to dashboard Cite

Scientific research papers must be published to visualize the results obtained by the experiments conducted as part of their research work. Every scientific article must have some experimental results. These results are a small step of the research work carried out by the researchers. All the experimentsbased research articles must follow some of the minimum structure when writing them. Most of the articles have utilized the standard format which is recognized by peer-reviewed journals. Some of the articles are not using the proper structure. This research work must discuss the basic requirements for the preparation of scientific research papers. Also, summarise some of the rules about how to prepare papers that are easily accepted by the indexed journals. The drawbacks and advantages are also discussed in this work to write experimental papers effectively and efficiently. A commonly acceptable format for writing the articles is specified and suggested based on their merits and demerits in this work.

show abstract

“…Two 16 independent reviews comparing rule-based and trained machine learning approaches reported 17 rule-based approaches to be superior in performance [2,3]. Another work reported that machine 18 learning models modestly outperformed rule-based classifiers with out-of-context training, yielding 19 further improvements with in-context training [4]. However, this work compares several machine 20 learning classifiers with one particular implementation of NegEx, that is not used in any of the 21 previous two studies, revealing superior rule-based performance, and does not consider it for 22 additional training in any context (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…It is reported as having a similar performance to ConText, with a 112 slightly higher recall. In an independent assessment [19], it was reported to perform extremely well 113 when extended with a richer negation vocabulary [19], outperforming other popular methods. 114…”

mentioning

confidence: 99%

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

Slater

Bradlow

Motti

et al. 2020

Preprint

View full text Add to dashboard Cite

AbstractBackgroundNegation detection is an important task in biomedical text mining. Particularly in clinical settings, it is of critical importance to determine whether findings mentioned in text are present or absent. Rule-based negation detection algorithms are a common approach to the task, and more recent investigations have resulted in the development of rule-based systems utilising the rich grammatical information afforded by typed dependency graphs. However, interacting with these complex representations inevitably necessitates complex rules, which are time-consuming to develop and do not generalise well. We hypothesise that a heuristic approach to determining negation via dependency graphs could offer a powerful alternative.ResultsWe describe and implement an algorithm for negation detection based on grammatical distance from a negatory construct in a typed dependency graph. To evaluate the algorithm, we develop two testing corpora comprised of sentences of clinical text extracted from the MIMIC-III database and documents related to hypertrophic cardiomyopathy patients routinely collected at University Hospitals Birmingham NHS trust. Gold-standard validation datasets were built by a combination of human annotation and examination of algorithm error. Finally, we compare the performance of our approach with four other rule-based algorithms on both gold-standard corpora.ConclusionsThe presented algorithm exhibits the best performance by f-measure over the MIMIC-III dataset, and a similar performance to the syntactic negation detection systems over the HCM dataset. It is also the fastest of the dependency-based negation systems. Our results show that dependency-based algorithms, utilising a single heuristic, can be powerful and stable methods for negation detection in clinical text, requiring minimal training and adaptation between datasets. While NegEx retains an extremely high performance in some cases, the presented approach may be more robust to more complex text descriptions. As such, it could present a drop-in replacement or augmentation for syntactic negation components in clinical text-mining pipelines, particularly for cases where adaptation and rule development is not required or possible.

show abstract

Evaluation of lexicon- and syntax-based negation detection algorithms using clinical text data

Cited by 3 publications

References 27 publications

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

Reviewer Perspective on the Essentials of Scientific Experiments Based Research Articles

A fast, accurate, and generalisable heuristic-based negation detection algorithm for clinical text

Contact Info

Product

Resources

About