OR29-02 Natural Language Processing of Radiology Reports Improves Identification of Patients with Fracture

Atypical femur fractures (AFF) are a rare but serious complication of long‐term bisphosphonate use. Although clearly defined by ASBMR criteria, a proportion of patients with AFFs may go unrecognized and the use of qualitative fracture criteria may lead to uncertainty in AFF diagnosis, with significant therapeutic implications. A score that rapidly and accurately identifies AFFs among subtrochanteric femur fractures using quantitative, measurable parameters is needed. In a retrospective cohort of 110 female patients presenting with AFFs or typical femur fractures (TFFs), multiple logistic regression and decision tree analysis were used to develop the Sydney AFF score. This score, based on demographic and femoral geometry variables, uses three dichotomized independent predictors and adds one point for each: (age ≤80 years) + (femoral neck width <37 mm) + (lateral cortical width at lesser trochanter ≥5 mm), (score, 0 to 3). In an independent validation set of 53 female patients at a different centre in Sydney, a score ≥2 demonstrated 73.3% sensitivity and 69.6% specificity for AFF (area under the receiver‐operating characteristic curve [AUC] 0.775, SE 0.063) and remained independently associated with AFF after adjustment for bisphosphonate use. The Sydney AFF score provides a quantitative means of flagging female patients with atraumatic femur fractures who have sustained an AFF as opposed to a TFF. This distinction has clear management implications and may augment current ASBMR diagnostic criteria. © 2021 American Society for Bone and Mineral Research (ASBMR).

show abstract

“…Similar artificial intelligence (AI)‐based systems for fracture identification have gained recent attention. ( 41 )…”

Section: Discussionmentioning

confidence: 99%

The Sydney AFF Score: A Simple Tool to Distinguish Females Presenting With Atypical Femur Fractures Versus Typical Femur Fractures

Crouch

Dhanekula

Burn

et al. 2020

Journal of Bone and Mineral Research

View full text Add to dashboard Cite

show abstract

“…Machine learning models have shown great promise in assessing and predicting mental disorders based on widely different datasets, for genetics, magnetic resonance imaging (MRI), electroencephalography (EEG), and clinical data (e.g., Allesøe et al, 2023). In clinical settings, a popular application of ML has been NLP, which has been used especially in electronic health records (e.g., Castro et al, 2015;Navarro et al, 2021), medical diagnosis (e.g., Kolanu et al, 2020) and social media text data mining (e.g., Levanti et al, 2023). Nevertheless, using such input data necessitates participants having access to, and being able to provide, a substantial amount of up-to-date medical records.…”

Section: Analyzing Mental Health With Nlp/ml Methodsmentioning

confidence: 99%

Question-Based Computational Language Approach Outperforms Rating Scales in Quantifying Emotional States

Sikström,

Valavičiūtė,

Kuusela

et al. 2023

Preprint

View full text Add to dashboard Cite

Psychological constructs are commonly quantiﬁed with closed-ended rating scales, however, recent advances in natural language processing (NLP) allow for quantiﬁcation of open-ended language responses with unprecedented accuracy. We demonstrate that specific open-ended question analyzed by natural language processing (NLP) shows higher accuracy in categorizing emotional states compared to traditional rating scales. One group of participants (N = 297) was asked to generate narratives related to four emotions; depression, anxiety, satisfaction, or harmony. The second group of participants (N = 434), read the narratives produced by the ﬁrst group. Both groups summarized the narratives in ﬁve descriptive words and rated the narratives on four rating scales commonly used to measure these emotions. The descriptive words were quantiﬁed by NLP methods and machine learning was used to categorize the responses into the corresponding emotional categories. The results showed a substantially higher number of accurate categorizations of the narratives based on descriptive words (64%) than on rating scales (44%), indicating that semantic measures have signiﬁcantly higher predictive accuracy than the corresponding four rating scales. These findings are important, as it contradicts the commonly held view that rating scales have higher accuracy in quantifying mental states than language-based measures.

show abstract

“…The model accuracy improves as more data become available, and the system can 'learn' or improve its performance on a set of tasks without external intervention or supervision. Recent developments in artificial intelligence have simplified the work of marketing professionals, clinicians, statisticians, and various analysts, and promising results have been observed in fields like web searches 11,12 , medical diagnosis 13 , targeted marketing 14 , and finance 15 . In clinical settings, a popular application of ML has been NLP, which has been used especially in electronic health records 16 .…”

Section: Introductionmentioning

confidence: 99%

Question-based computational language assessment shows higher validity than rating scales in the categorization of emotional state narratives

Sikström

Kuusela

et al. 2022

Preprint

View full text Add to dashboard Cite

Psychological constructs are commonly quantified with closed-ended rating scales, however, recent advances in natural language processing (NLP) have been shown to allow for the quantification of open-ended language responses with unprecedented accuracy. Here, we showed that emotional events analysed by NLP show higher accuracy in categorizing emotional states than rating scales. One group of participants (N = 297) was asked to generate narratives related to four emotions – depression, anxiety, satisfaction, or harmony and to rate them on four standardized rating scales (i.e., PHQ-9, GAD-7, SWLS, and HILS). The second group of participants (N = 400 controls and N = 34 healthcare professionals), read the narratives produced by the first group, summarized them in five descriptive keywords, and rated the narratives on the same four emotional states. The descriptive words were quantified by NLP methods, and machine learning was used to categorize the responses into the corresponding emotional categories. The results showed a substantially more accurate categorization of the narratives based on descriptive words (64 %) than on rating scales (44 %), indicating that semantic measures have significantly higher predictive accuracy than the corresponding four rating scales. In conclusion, our results show that semantic responses can have a higher validity in categorizing narratives of emotions than validated Likert scales.

show abstract

OR29-02 Natural Language Processing of Radiology Reports Improves Identification of Patients with Fracture

Cited by 6 publications

References 0 publications

The Sydney AFF Score: A Simple Tool to Distinguish Females Presenting With Atypical Femur Fractures Versus Typical Femur Fractures

The Sydney AFF Score: A Simple Tool to Distinguish Females Presenting With Atypical Femur Fractures Versus Typical Femur Fractures

Question-Based Computational Language Approach Outperforms Rating Scales in Quantifying Emotional States

Question-based computational language assessment shows higher validity than rating scales in the categorization of emotional state narratives

Contact Info

Product

Resources

About