SlideShare a Scribd company logo
1 of 14
1
Pharmabroad:
A Companion Chatbot for Identifying
Pharmaceutical Products When Traveling
Abroad
08/01/2020 – University of Surrey, UK
Sammarco Matteo
Ruf Boris (Axa), Sammarco Matteo (Axa), Jonathan Aigrain (Axa),
Marcin Detyniecki (Axa, Polish Academy of Science)
2
A lost in translation story
An Italian tourist in France
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
3
A lost in translation story
A French tourist in Italy
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
4
More complex situations
Different language and alphabet
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
5
Contribution
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling
Abroad
Match local pharmaceutical products with travelers’ home country products
Pervasive and inexpensive
 Smartphone ready
Simplified interaction (chatbot)
 Manage stressful situations
 No need to write product name in different alphabet (OCR)
World-wide scalability
6
System design
Interface/Dialog flow
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Interface/
Dialog flow
Core service
Matching
algorithm
7
Medical box
image I
OCR(I)
Host country DB lookup
Home country DB lookup
Geolocation
Host
country
DB
available
?
No
Yes
medicinename ! molecule
molecule ! medicinename
T(OCR(I))
System design
Core service
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Interface/
Dialog flow
Core service
Matching
algorithm
8
System design
Matching algorithm
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Interface/
Dialog flow
Core service
Matching
algorithm
9
Matching algorithm
Local drug DB available
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Candidate 1
Candidate 2
Local
drug
DB
(L)
OCR
Home
drug
DB
(L)
Levenshtein
distance < 𝜏
Result:
molecule
Looking for
drug name
10
Matching algorithm
Local drug DB not available
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Home
drug
DB
(L)
OCR
Molecule
translation
Levenshtein
distance < 𝜏
11
Matching results
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Local drug DB available Local drug DB not available
OCR mistake OCR mistake
Smaller fonts
Translation error
12
Implementation
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
OCR: OCR.space, Microsoft Azure Vision Service
NLP: Microsoft Azure Bot Service
Translation: Microsoft Translator Text API
Drug database:
 ChEMBL,
 AIFA (Agenzia Italiana del Farmaco),
 FDA (Food and Drug Administration)
13
Conclusion
Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
Black box service
Good balance between precision and recall (F1-score = 0.87) with low Levenshtein distance
(𝜏 = 3) when the local country drug DB is available
Getting the molecule directly is more challenging (F1-score = 0.66)
Embed vision and translation services inside the mobile application (no network connection
needed)
14
Thank you

More Related Content

Recently uploaded

Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
panagenda
 

Recently uploaded (20)

Portal Kombat : extension du réseau de propagande russe
Portal Kombat : extension du réseau de propagande russePortal Kombat : extension du réseau de propagande russe
Portal Kombat : extension du réseau de propagande russe
 
How we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdfHow we scaled to 80K users by doing nothing!.pdf
How we scaled to 80K users by doing nothing!.pdf
 
Overview of Hyperledger Foundation
Overview of Hyperledger FoundationOverview of Hyperledger Foundation
Overview of Hyperledger Foundation
 
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
 
ECS 2024 Teams Premium - Pretty Secure
ECS 2024   Teams Premium - Pretty SecureECS 2024   Teams Premium - Pretty Secure
ECS 2024 Teams Premium - Pretty Secure
 
State of the Smart Building Startup Landscape 2024!
State of the Smart Building Startup Landscape 2024!State of the Smart Building Startup Landscape 2024!
State of the Smart Building Startup Landscape 2024!
 
Designing for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at ComcastDesigning for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at Comcast
 
Extensible Python: Robustness through Addition - PyCon 2024
Extensible Python: Robustness through Addition - PyCon 2024Extensible Python: Robustness through Addition - PyCon 2024
Extensible Python: Robustness through Addition - PyCon 2024
 
BT & Neo4j _ How Knowledge Graphs help BT deliver Digital Transformation.pptx
BT & Neo4j _ How Knowledge Graphs help BT deliver Digital Transformation.pptxBT & Neo4j _ How Knowledge Graphs help BT deliver Digital Transformation.pptx
BT & Neo4j _ How Knowledge Graphs help BT deliver Digital Transformation.pptx
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
 
Intro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераIntro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджера
 
A Business-Centric Approach to Design System Strategy
A Business-Centric Approach to Design System StrategyA Business-Centric Approach to Design System Strategy
A Business-Centric Approach to Design System Strategy
 
Easier, Faster, and More Powerful – Notes Document Properties Reimagined
Easier, Faster, and More Powerful – Notes Document Properties ReimaginedEasier, Faster, and More Powerful – Notes Document Properties Reimagined
Easier, Faster, and More Powerful – Notes Document Properties Reimagined
 
WebAssembly is Key to Better LLM Performance
WebAssembly is Key to Better LLM PerformanceWebAssembly is Key to Better LLM Performance
WebAssembly is Key to Better LLM Performance
 
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfLinux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
 
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
Secure Zero Touch enabled Edge compute with Dell NativeEdge via FDO _ Brad at...
 
The Value of Certifying Products for FDO _ Paul at FIDO Alliance.pdf
The Value of Certifying Products for FDO _ Paul at FIDO Alliance.pdfThe Value of Certifying Products for FDO _ Paul at FIDO Alliance.pdf
The Value of Certifying Products for FDO _ Paul at FIDO Alliance.pdf
 
TopCryptoSupers 12thReport OrionX May2024
TopCryptoSupers 12thReport OrionX May2024TopCryptoSupers 12thReport OrionX May2024
TopCryptoSupers 12thReport OrionX May2024
 
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
Easier, Faster, and More Powerful – Alles Neu macht der Mai -Wir durchleuchte...
 
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
TEST BANK For, Information Technology Project Management 9th Edition Kathy Sc...
 

Featured

Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
Kurio // The Social Media Age(ncy)
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Saba Software
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming Language
Simplilearn
 

Featured (20)

How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024How to Prepare For a Successful Job Search for 2024
How to Prepare For a Successful Job Search for 2024
 
Social Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie InsightsSocial Media Marketing Trends 2024 // The Global Indie Insights
Social Media Marketing Trends 2024 // The Global Indie Insights
 
Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024Trends In Paid Search: Navigating The Digital Landscape In 2024
Trends In Paid Search: Navigating The Digital Landscape In 2024
 
5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary5 Public speaking tips from TED - Visualized summary
5 Public speaking tips from TED - Visualized summary
 
ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd ChatGPT and the Future of Work - Clark Boyd
ChatGPT and the Future of Work - Clark Boyd
 
Getting into the tech field. what next
Getting into the tech field. what next Getting into the tech field. what next
Getting into the tech field. what next
 
Google's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search IntentGoogle's Just Not That Into You: Understanding Core Updates & Search Intent
Google's Just Not That Into You: Understanding Core Updates & Search Intent
 
How to have difficult conversations
How to have difficult conversations How to have difficult conversations
How to have difficult conversations
 
Introduction to Data Science
Introduction to Data ScienceIntroduction to Data Science
Introduction to Data Science
 
Time Management & Productivity - Best Practices
Time Management & Productivity -  Best PracticesTime Management & Productivity -  Best Practices
Time Management & Productivity - Best Practices
 
The six step guide to practical project management
The six step guide to practical project managementThe six step guide to practical project management
The six step guide to practical project management
 
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
Beginners Guide to TikTok for Search - Rachel Pearson - We are Tilt __ Bright...
 
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
Unlocking the Power of ChatGPT and AI in Testing - A Real-World Look, present...
 
12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work12 Ways to Increase Your Influence at Work
12 Ways to Increase Your Influence at Work
 
ChatGPT webinar slides
ChatGPT webinar slidesChatGPT webinar slides
ChatGPT webinar slides
 
More than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike RoutesMore than Just Lines on a Map: Best Practices for U.S Bike Routes
More than Just Lines on a Map: Best Practices for U.S Bike Routes
 
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
Ride the Storm: Navigating Through Unstable Periods / Katerina Rudko (Belka G...
 
Barbie - Brand Strategy Presentation
Barbie - Brand Strategy PresentationBarbie - Brand Strategy Presentation
Barbie - Brand Strategy Presentation
 
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them wellGood Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
Good Stuff Happens in 1:1 Meetings: Why you need them and how to do them well
 
Introduction to C Programming Language
Introduction to C Programming LanguageIntroduction to C Programming Language
Introduction to C Programming Language
 

Pharmabroad a companion chatbot for identifying pharmaceutical products when traveling abroad

  • 1. 1 Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad 08/01/2020 – University of Surrey, UK Sammarco Matteo Ruf Boris (Axa), Sammarco Matteo (Axa), Jonathan Aigrain (Axa), Marcin Detyniecki (Axa, Polish Academy of Science)
  • 2. 2 A lost in translation story An Italian tourist in France Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
  • 3. 3 A lost in translation story A French tourist in Italy Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
  • 4. 4 More complex situations Different language and alphabet Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad
  • 5. 5 Contribution Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Match local pharmaceutical products with travelers’ home country products Pervasive and inexpensive  Smartphone ready Simplified interaction (chatbot)  Manage stressful situations  No need to write product name in different alphabet (OCR) World-wide scalability
  • 6. 6 System design Interface/Dialog flow Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Interface/ Dialog flow Core service Matching algorithm
  • 7. 7 Medical box image I OCR(I) Host country DB lookup Home country DB lookup Geolocation Host country DB available ? No Yes medicinename ! molecule molecule ! medicinename T(OCR(I)) System design Core service Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Interface/ Dialog flow Core service Matching algorithm
  • 8. 8 System design Matching algorithm Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Interface/ Dialog flow Core service Matching algorithm
  • 9. 9 Matching algorithm Local drug DB available Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Candidate 1 Candidate 2 Local drug DB (L) OCR Home drug DB (L) Levenshtein distance < 𝜏 Result: molecule Looking for drug name
  • 10. 10 Matching algorithm Local drug DB not available Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Home drug DB (L) OCR Molecule translation Levenshtein distance < 𝜏
  • 11. 11 Matching results Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Local drug DB available Local drug DB not available OCR mistake OCR mistake Smaller fonts Translation error
  • 12. 12 Implementation Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad OCR: OCR.space, Microsoft Azure Vision Service NLP: Microsoft Azure Bot Service Translation: Microsoft Translator Text API Drug database:  ChEMBL,  AIFA (Agenzia Italiana del Farmaco),  FDA (Food and Drug Administration)
  • 13. 13 Conclusion Pharmabroad: A Companion Chatbot for Identifying Pharmaceutical Products When Traveling Abroad Black box service Good balance between precision and recall (F1-score = 0.87) with low Levenshtein distance (𝜏 = 3) when the local country drug DB is available Getting the molecule directly is more challenging (F1-score = 0.66) Embed vision and translation services inside the mobile application (no network connection needed)

Editor's Notes

  1. Hello, I’m glad to present you today Pharmabroad, whici is a system with a chatbot interface to identify pharmaceutical products when traveling abroad.
  2. This work began from a personal experience: 10 years ago I moved from Italy to France and I packed in my bag some basic medicaments. But when traveling you remain long enough abroad and you run out of personal stock or you forgot to bring something with you or you just didn’t include it in the travel check list, you need to buy some local medical products. In France, eventually, you’ll come across a medicament called « Doliprane » and I had to take some pills of Doliprane too. So I discovered which is simply made by paracetamol as molecule and used maily for fever treatment, or headache or just some pain. It can by bought without medical presciption.
  3. Reading further the warning label, for the chemical composition, side effects, posology and so on, I discovered that the Doliprane is the equivalent of what, at contrary, a French turist staying in Italy can find as Tachipirina. This is a kind of toy case scenario, as the principal molecule was written clearly on the box, but in many cases it’s not so simple.
  4. In fact, we can travel in contries where the spoken and written language is different, so that the warning label could be impossible to decrypt. Even the alphabet could be different and it would result impossible for us even to write the name to google it. In addition, when you need some medicines it’s supposed you have some pain or you are sick and it’s a stressful. moment in general.
  5. This is why we designed Pharmabroad: to match pharmaceutical products from the host country to the home country products. As requirements, we wanted it to be pervasive and cheap: so basically it’s a smartphone application With a simplified interaction: thus we adopted the a chatbot paradigm with a minimal interaction with the user, without even the need to write the medicament name And world-wide scalable.
  6. More in detail, I divide the system in 3 parts: The HM interface with the dialog flow The core service The matching algo This is the higl level interaction schema… Other intents: I lost my luggage, am I insured for this…
  7. RESULT: The algorithm can return either the right correspondent drug, or ”nothing” which is a False Negative, or a different drug with is a False Positive.
  8. Results are achieved using only one image for each pharmaceutical box and figures used in the experiments have a much lower resolution wrt usual smartphones.
  9. Experiments show that we can get a good balance between precision and recall for medicine name extraction from drug boxes images (F1 score = 0 . 87) keeping a low Levenshtein distance ( τ = 3)