Leveraging Text Classification Strategies for Clinical and Public Health Applications

Leveraging Text Classification Strategies for Clinical and Public Health Applications

Karin M. Verspoor

@[email protected] University of MelbourneMelbourne, Victoria, Australia

January 2016, Qatar Computing Research Institute

(clinical) Data everywhere

• Electronic health records– Patient demographics and biometrics– Laboratory test results– Clinical notes

• Radiology and pathology– Images: X-ray, MRI and PET Scans– (Synoptic) Reports

• Databases– Health Service reporting– National Prescribing Service– Registry data, Births and Deaths– Medicare/insurance claim data etc…..

Don’t forget unstructured data!

• About 80% of clinical information is in textual form– ED triage notes– Clinical progress notes– Radiology and Pathology reports– GP and specialist letters– Discharge summaries

• Published Literature– Clinical Trials– Molecular-level studies

• and … social media text!

How is text used in medicine?

• Direct analysis of clinical records– Information retrieval for clinical trials– Syndromic surveillance– Hospital Services Research– Clinical Decision Support– Pharmacovigilance

• Literature mining– Evidence-based medicine– Systematic Reviews

Evidence from EHRs

Mining electronic health records: towards better research applications and clinical carePeter B. Jensen, Lars J. Jensen & Søren Brunak, Nature Reviews Genetics 13, 395-405 (2012)doi:10.1038/nrg3208

Pharmacovigilance from EHRs

Mining of clinical records to identify adverse drug events

Estimated >90% of adverse events do not appear in coded data

6LePendu et al. (2013) “Pharmacovigilance Using Clinical Notes” Clinical Pharmacology & Therapeutics 93(6), 547–555; doi: 10.1038/clpt.2013.47

… from social media

Pacific Symposium on Biocomputing Shared Task on Social Media Mining

Classification of tweets: mention an Adverse Drug Reaction?

ADR classified@NAME Q makes me hungry. Olanzapine made me want to eat my own arm!

Non-ADR classifiedI couldnt be a chef without nicotine and caffeine

OutlineProblem settingApproach and Results

EHR disease classification

Kocbek et al (2015) Evaluating classification power of linked admission data sources with text mining; Proceedings of the Scientific Stream at Big Data in Health Analytics 2015 (BigData 2015).

ICD classification of EHR data

• We address the task of detecting clinical records in a large record system corresponding to a given diagnosis of interest, based on text analysis

• We focus on lung cancer records for a pilot study

• We developed a system that classifies each admission as positive or negative for lung cancer

• Not as simple as looking for “lung cancer” or synonyms in the EHRs!

Kocbek et al (2015) HISA Big Data conference. http://ceur-ws.org/Vol-1468/bd2015_kocbek.pdf

Alfred REASON platformKocbek et al. Big Data 2015, Sydney

• 15+ years of data from. • 171,000+ updates each day.• 62.4 million updates per annum.

Radiology question

50yo complaining of left shoulder pain. Tender generally. Difficulty abducting the shoulder past 45 degrees. Home on HITH tomorrow - either inpatient or outpatient please

Task

Radiology report

Mobile Chest performed on 02-JUN-2012 at 08:27 AM: The nasogastric tube has its tip in the stomach. The tracheostomy is seen at T2 level. ….

Pathology report

Urine Culture Acc No: 12-183-0731Source: Urine ------------ URINE MICROSCOPY (PHASE CONTRAST) ------------- Leucocytes x10^6/L (Ref <10).... <10 Erythrocytes x10^6/L (Ref <10).. <10.......

Additional data

Age: 50Date of admission: Jun/12Gender: FCountry: …

Admission

ICD-10 code

Data Characteristics

• Extracted data for 2 financial years, 2012-2014:– 150,521 admissions, – 40,800 radiology reports with associated

question,– 20,872 pathology reports,– 121,700 additional data entries (demographics,

hospital admission info).• Admissions are associated to ICD-10 codes:

– Used as ground truth– ICD-10 code C34.*; positive cases for lung

cancer– 496 such positive admissions– an additional 496 non-lung cancer submissions

randomly subsampled as negatives

OutlineProblem settingApproach and Results

EHR disease classification

Research Question

• Most previous TM applications use a single textual data source from the EHR despite a diversity of potential data

• What is the impact of using more than one textual data source for the EHR classification task?– Considering different text sources;– and including patient (structured) meta-data?

Methods

Radiology reports

Machine learning algorithm (SVM)

Textual and other features

Biomedical knowledge

sources Language processing

ClassificationModel

Additional data

Pathology reports

Radiology questions

REASON sources

Text Processing

• Medical terminology recognition and normalisation using MetaMap

• NegEx to detect negation and negation scope

The nasogastric tube has its tip in the stomach.Meta Candidates (Total=9; Excluded=0; Pruned=0; Remaining=9)1000 C0085678:Nasogastric tube [Medical Device]1000 C0812428:Nasogastric tube (Nasogastric tube procedures) [Therapeutic Procedure] 861 C0175730:Tube (biomedical tube device) [Medical Device]861 C0694637:Nasogastric (Nasogastric Route of Drug Administration) [Functional Concept] 861 C1547937:Tube NOS (Specimen Source Codes - Tube) [Intellectual Product]861 C1561954:tube [Conceptual Entity]861 C1704730:TUBE (Packaging Tube) [Medical Device]861 C1704731:Tube (Tube Device Component) [Medical Device]861 C3282907:Nasogastric [Body Location or Region]

Meta Mapping (1000):1000 C0085678:Nasogastric tube [Medical Device]

Features

Texts• bag of (MetaMap) phrases

– separate feature for Positive/Negative context– experimented with keeping phrases separated

according to source, or merging across sources

Patient meta-data• demographic data (gender, age, ethnic origin,

country, language, marital status, religion, and death date)

• hospital-related admission data (hospital code, admission date and time, discharge date and time, length of stay, reason for admission, admission unit, discharge unit, admission type, source, destination and criteria)

Experimental setting

• Heavily skewed data: undersampling of negatives

• 10-fold cross validation• Support Vector Machine (Weka)

Results: Lung Cancer

1 2 3 40.870

0.885

0.900

0.915

0.930

0.873

0.901

radiology reports + 1 data source(F-Score)

radiology question pathology report additional data

1 2 3 40.870

0.885

0.900

0.915

0.930

0.873

0.901

0.917

radiology reports + 2 additional data sources

(F-score)

radiology question pathology reports additional data


1 2 3 40.870

0.885

0.900

0.915

0.930

0.873

0.901

0.917

0.93

F-Score using 4 data sources

radiology question pathology reports additional data


Discussion

• More data sources lead to better performance

• The classifier with the highest performance was built using features from all four data sources

• Merging sources into aggregate features better

• Not all improvements are significant:– Radiology question and metadata add clear

value– Pathology reports does not

• Not all admissions had a pathology report associated with them.

Case study 1: Conclusions

• We built a text mining system for detecting lung cancer admissions using machine learning methods.

• Our results show more effective systems can generally be built by including multiple linked data sources.

• Work in progress:– Other diseases– Imbalanced datasets– Feature engineeringand selection

1 2 3 40.820

0.830

0.840

0.850

0.860

0.870

0.880

0.890

0.900

0.910

0.920

0.893

Breast cancer

OutlineDOD with TwitterEmotion classificationDOD signal 1: Tweet emotion shiftDOD signal 2: Tweet lexical shift

Disease Outbreak Detection

Ofoghi et al (2016) Towards early discovery of salient health threats: A social media emotion classification technique; Pacific Symposium on Biocomputing.

Twitter for Outbreak Detection

Assumptions• People tweet about diseases in the context

of emerging outbreaks• Twitter can provide an “early warning” of an

outbreak

“Tweets started to rise in Nigeria 3-7 days prior to the official announcement of the first probable Ebola case. The topics discussed in tweets include risk factors, prevention education, disease trends, and compassion.”

Amer J Infection Control (2015)

“Early warning” tweets

Ebola on Twitter

28

Twitter for Outbreak Detection

Strategy• Trends: counting of (hashtag, term)

frequencies• Coupled with geographic origin of tweets• Sentiment or content analyis

Challenges• High volume of (mostly irrelevant) tweets• Hashtags alone may not be adequate• A mention of a disease does not necessarily

indicate an active case

Many reasons to mention Ebola

DOD with Twitter | Previous Work

31

Is there a local emergent threat?

Can we use shifts in emotional and lexical content of tweets

to detect a disease outbreak?

A sliding window model

Ebola event/background data

Dataset Date (±7) pre-corpus post-corpus#tweets |vocab| #tweets |vocab|

ebola-event-1 29-Dec-14 73 204 337 906

ebola-event-2 31-Jan-15 165 700 90 417

ebola-background 16-12-14 429 1453 340 1208

Emotion classes

• ECs: Ekman’s six basic emotions plus …– News-related– Criticism– Sarcasm

https://www.behance.net/gallery/6-Basic-Emotions/930168

Sarcasticatsign atsign think I got Ebola there two minutes ago

News-relatedatsign Another 4 American Ebola workers flown back to USA for monitoring..



Emotion classifier data

• Data: collection– Twitter API– Second half of March 2015– Total of 12,101 tweets– Contained “ebola” or “#ebola”– 4,405 tweets remained after some filtering…– Amazon’s Mechanical Turk was used to label

tweets

Lexicon-Based Classification

• Created an emotion vocabulary– Profile of Mood States (POMS)– FrameNet– Existing “feelings list”– Wikipedia

• Vector space model– Binary vector per emotion– Binary vector per tweet– Cosine Similarity emotion vs tweet

1

2

3 anxious

4

5

6

7 affronted

8

9

497

498

499 :-|

.

.

.

https://bitbucket.org/readbiomed/socialsurveillance







Emotion class distribution

Classes Dataset p-value6 emotions ebola-event-1 0.004*

ebola-event-2 0.002*ebola-backgr. 0.259

6 emotions + 3 add’l ebola-event-1 0.009*ebola-event-2 0.007*ebola-backgr. 0.079

paired t-test, pre- and post-event windows; * Statistically significant at 5% level

Jensen-Shannon divergence

Class ebola-event-1

ebola-event-2

ebola-bg.

Sarcasm 0.0227 0.0032 0.1365News-rel. 0.0226 0.0001 0.0074Anger 0.0572 0.0382 0.0169Criticism 0.0180 0.0056 0.0060Surprise 0.1161 0.0220 0.0023Fear 0.0768 0.0813 0.0913Happiness 0.0444 0.0415 0.0064Disgust 0.0604 0.0025 0.0044Sadness 0.0023 0.0322 0.0060AVERAGE 0.0467 0.0252 0.0308

Big differences compared with background, in both e1 and e2

Lexical shift analysis

Within-corpus analysis:

Cross-corpus analysis:

Term freq changes: Event 1

Term freq changes: Background

Case study 2: Conclusions

• We introduced an Ebola tweet-based emotion classifier.

• There are statistically significant differences in the distribution of emotion classes and lexical items in tweets preceding and following a salient emergent health threat.

• This effect does not occur in a neutral background collection.

Proposal:• Disease outbreak detection can be

supported with monitoring of tweets using a sliding window model that tests for such distributional changes

Conclusions

• There are myriad problems in the clinical context where unstructured data can be leveraged to good effect

• Text classification is one tool that can be drawn on to make use of this unstructured data

• Heterogeneous data integration is also important

• Challenges exist in – Terminology– Skewed data– Missing data

Acknowledgements

• Amazon Mechanical Turkers• James McCaw, Melbourne School of Population and

Global Health

Bahador Ofoghi

Lawrence Cavedon

Simon Kocbek

Thank you!

ML-Based Classification

• MALLET Naïve Bayes• Features

– bag of words[+lem,-lem]– Lexicon-based similarity– emotion vocabulary– emoticons– punctuation– (Stanford) sentiment

KL-Divergence, full vocabulary

Emotion-level distribution

KL-divergence (pre- vs. post-event, post- vs. pre-)

P(x) and Q(x) represent probability of positive and negative emotion classesin the respective corpora

Top lexically distinct items

Log Likelihood analysis

Leveraging Text Classification Strategies for Clinical and Public Health Applications

Health & Medicine

Transcript of Leveraging Text Classification Strategies for Clinical and Public Health Applications