/
NAACCR Cancer Informatics HackathonTeam NLP CommandersJune 2018 NAACCR Cancer Informatics HackathonTeam NLP CommandersJune 2018

NAACCR Cancer Informatics HackathonTeam NLP CommandersJune 2018 - PDF document

alyssa
alyssa . @alyssa
Follow
344 views
Uploaded On 2021-08-26

NAACCR Cancer Informatics HackathonTeam NLP CommandersJune 2018 - PPT Presentation

Our TeamDr Patrick McNeillieUniversity of North Carolina 2005National Institute of Technology JaipurAakash BhatiaAditya ChindhadeKedar DabhadkarDr Jeffrey BondMohit ThakurInstitute of Chemical Technol ID: 872051

polyp colon cancer 148 colon polyp 148 cancer 147 123 sigmoid ascending final patient date pathology examination submitted 5551234

Share:

Link:

Embed:

Download Presentation from below link

Download Pdf The PPT/PDF document "NAACCR Cancer Informatics HackathonTeam ..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

1 NAACCR Cancer Informatics HackathonTeam:
NAACCR Cancer Informatics HackathonTeam: NLP Commanders,June 2018 Our Team Dr. Patrick McNeillie University of North Carolina 2005 National Institute of Technology Jaipur Aakash Bhatia Aditya Chindhade Kedar Dabhadkar Dr. Jeffrey Bond Mohit Thakur Institute of Chemical Technology, Mumbai Wisconsin Cancer Reporting SystemPhD in Biophysics, Universi

2 ty of Rochester OutlineChallenge Introdu
ty of Rochester OutlineChallenge IntroductionApproachBaseline ModelFinal ModelConclusionFuture Work 1. Challenge Introduction Problem Statement Expert’s notes Code BREASTCOLONPROSTRATEOTHER UnformattedGenerated Sample Pathology Report (from Orchard Pathology Laboratories)Patient Name: Patient, John. M,Age 34 | DOB: 4/12/1979 Phone: (123) 5551

3 234. EMR: (123) 5551234., PHYSICIAN INFO
234. EMR: (123) 5551234., PHYSICIAN INFORMATION: James Provider, MD ABC Medical 400 Royal Drive Anytown USA 12345 Phone: (123) 5554321 Fax: (999) 4322., XODREPORT DATE: 2/17/2013 TAT: [26 hours], Specimen: 2 cm polyp ascending colon 2 mm polyp in sigmoid colon Clinical History: Screening colonoscopy. Maternal hx of adenocarcinoma of colon age 57 G

4 ross Examination A. The first container
ross Examination A. The first container is labeled “ascending colon.” It contains a polypoid piece of tan mucosal tissue measuring 2.0 cm in greatest dimension. The polyp margin is inked, sectioned, and submitted in cassettes Al and A2. B. The second container is labeled “sigmoid colon.” It contains one piece of light tan mucos

5 al tissue 0.2 cm in greatest dimension.
al tissue 0.2 cm in greatest dimension. Entirely submitted in cassette B. Microscopic Examination Microscopic Examination performed supportive of the Final DiagnosisXODA, FINAL DIAGNOSIS A. Ascending Colon SESSILE SERRATED ADENOMA (POLYP) WITH LOWGRADE ADENOMATOUS DYSPLASIA. B. Sigmoid Colon TUBULAR ADENOMA COMMENT: XODPatients with sessile serrat

6 ed adenomas, especially with cytologic d
ed adenomas, especially with cytologic dysplasia, are at increased risk for the development of adenocarcinoma showing microsatellite instability. This progression may occur at a more rapid rate than with traditional adenomas. Complete endoscopic excision is recommended if clinically appropriate. If unresectable, repeat colonoscopy at a shortened i

7 nterval (1 year), with sampling ofsuspic
nterval (1 year), with sampling ofsuspicious areas or surgical resection possibly warrantedXODAACCESSION NUMBER 12XX0002,COLLECTION DATE: 2/15/2013 RECEIVED DATE: 2/15/2013 http://www.orchardsoft.com/files/reports/OrchardPathologyPatientReportExamples.pdf FormattedGenerated Sample Pathology Report (from Orchard Pathology Laboratories)Patient Name:

8 Patient, John. | M | DOB: 4/12/1979 | P
Patient, John. | M | DOB: 4/12/1979 | Patient ID :543216 | Phone: (123) 5551234 | EMR: (123) 5551234 Physician InformationJames Provider, MD | ABC Medical 400 Royal Drive Anytown USA 12345 | Phone: (123) 5554321 | Fax: (999) 5554322Final Diagnosis:A. Ascending Colon: SESSILE SERRATED ADENOMA (POLYP) WITH LOWGRADE ADENOMATOUS DYSPLASIAB. Sigmoid

9 Colon: TUBULAR ADENOMAComment:Patients w
Colon: TUBULAR ADENOMAComment:Patients with sessile serrated adenomas, especially with cytologic dysplasia, are at increased risk for the development of adenocarcinoma showing microsatellite instability. This progression may occur at a more rapid rate than with traditional adenomas. Complete endoscopic excision is recommended if clinically appropr

10 iate. If unresectable, repeat colonoscop
iate. If unresectable, repeat colonoscopy at a shortened interval (1 year), with sampling of suspicious areas or surgical resection possibly warranted.Accession Number: 12XX0002Collection Date:2/15/2013Received Date: 2/15/2013Report Date: 2/17/2013 TAT: [26 hours]Specimen: 2 cm polyp ascending colon 2 mm polyp in sigmoid colonClinical History: Scr

11 eening colonoscopy. Maternal hx of adeno
eening colonoscopy. Maternal hx of adenocarcinoma of colon age 57Gross ExaminationA. The first container is labeled “ascending colon.” It contains a polypoid piece of tan mucosal tissue measuring 2.0 cm in greatest dimension. The polyp margin is inked, sectioned, and submitted in cassettes Al and A2.B. The second container is labeled 

12 47;sigmoid colon.” It contains one
47;sigmoid colon.” It contains one piece of light tan mucosal tissue 0.2 cm in greatest dimension. Entirely submitted in cassette B.Microscopic Examination:Microscopic Examination performed supportive of the Final Diagnosis 2. Approach BAGOFWORDSSolution: Make a special word dictionary! All words Database Medical Expertise New dictionary of w

13 ords! 3. Baseline ModelCounter 3. Baseli
ords! 3. Baseline ModelCounter 3. Baseline ModelcounterCount the occurrences of 4 keywords: Prostate, Lung, Cancer, Breastin records.Classify the site based on the highest occurring keyword 3. Baseline Model ResultsF1 MACRO: 0.86078MODEL ACCURACY:85.76% Confusion matrix Final ModelNaive Bayes + SVM + Random Forests 4. Final ModelNaive Bayes + SVM

14 + Random Forests Naïve Bayes SVM Random
+ Random Forests Naïve Bayes SVM RandomForests 4. Final Model Results F1 MACRO: 0.936MODEL ACCURACY:Confusion matrix 5. Conclusion PROSTATEBREASTLUNGCOLONOTHER 6. Future work Challenges of natural language processingChallengeExampleNegationNo evidence of malignancy” in support of an OTHER classification.Ambiguity with respect to subjectA pat

15 hological observation may refer to a his
hological observation may refer to a historical sampleA LUNG cancer case has the phrase “cancer of the colon” because “the patient has a history of”.One pathology report may describe more than one sampleNo evidence of malignancy” occurs in a report of a cancer case because it refers to a sample from the tumor margin.Statis

16 tical sample sizeThe ‘OTHER’ c
tical sample sizeThe ‘OTHER’ class is a union of very different classes. The OTHER class comprises small numbers of samples representing noncancer as well as cancer of the blood, skin, stomach, etc.Latent cross classificationStochastic independence in the sampleThe identity of the registry may be associated with both SITE and usage (conf