full transcript

From the Ted Talk by Alicia Chong Rodriguez: A smart bra for better heart health

Unscramble the Blue Letters

When I was doing cardiovascular research at MIT, I had access to huge data sets. Realizing that women were one of the largest srbugoups ueeesepntrrndred was eye-opening. In fact, wmoen were basically excluded from cardiovascular cilcianl trials until the NIH mandated inclusion in 1993. This is why existing tloehociengs and therapies often fall short, because most of them have been designed using data primarily from male animals and men. And as artificial intelligence hleps turbocharge digital health, there's a danger that aohgtrlims mostly trained with male data and biases will actually perpetuate the problem.

Open Cloze

When I was doing cardiovascular research at MIT, I had access to huge data sets. Realizing that women were one of the largest _________ ________________ was eye-opening. In fact, _____ were basically excluded from cardiovascular ________ trials until the NIH mandated inclusion in 1993. This is why existing ____________ and therapies often fall short, because most of them have been designed using data primarily from male animals and men. And as artificial intelligence _____ turbocharge digital health, there's a danger that __________ mostly trained with male data and biases will actually perpetuate the problem.

Solution

  1. women
  2. subgroups
  3. helps
  4. clinical
  5. underrepresented
  6. technologies
  7. algorithms

Original Text

When I was doing cardiovascular research at MIT, I had access to huge data sets. Realizing that women were one of the largest subgroups underrepresented was eye-opening. In fact, women were basically excluded from cardiovascular clinical trials until the NIH mandated inclusion in 1993. This is why existing technologies and therapies often fall short, because most of them have been designed using data primarily from male animals and men. And as artificial intelligence helps turbocharge digital health, there's a danger that algorithms mostly trained with male data and biases will actually perpetuate the problem.

Frequently Occurring Word Combinations

ngrams of length 2

collocation frequency
heart attack 2
heart disease 2
medical devices 2
clinical trials 2
digital biomarkers 2

Important Words

  1. access
  2. algorithms
  3. animals
  4. artificial
  5. basically
  6. biases
  7. cardiovascular
  8. clinical
  9. danger
  10. data
  11. designed
  12. digital
  13. excluded
  14. existing
  15. fact
  16. fall
  17. health
  18. helps
  19. huge
  20. inclusion
  21. intelligence
  22. largest
  23. male
  24. mandated
  25. men
  26. mit
  27. nih
  28. perpetuate
  29. primarily
  30. problem
  31. realizing
  32. research
  33. sets
  34. short
  35. subgroups
  36. technologies
  37. therapies
  38. trained
  39. trials
  40. turbocharge
  41. underrepresented
  42. women