Background
Verbal autopsies
Metrics
Previous work
Methods
Data
Acute respiratory infections |
Diarrhea |
Pulmonary Tuberculosis |
Other and unspecified infections |
Neoplasms |
Nutrition |
Cardiovascular disease |
Chronic respiratory disease |
Liver cirrhosis |
Other non-communicable diseases |
Road and transport injuries |
Other injuries |
Ill-defined |
Suicide |
Maternal |
Prematurity/low birth weight |
Neonatal infections (not including tetanus) |
Birth asphyxia/trauma |
Ill-defined or cause unknown |
Other (all other ICDs not included in above) |
MDS | RCT | MDS+RCT | Agincourt | |
---|---|---|---|---|
Adult records (15–69 years) | 9,207 | 5,105 | 14,312 | 8,151 |
Child records (29 days–14 years) | 1,717 | 255 | 1,972 | 1,674 |
Neonatal records (<29 days) | 451 | 170 | 621 | 197 |
Region | India | India (Gujarat, Punjab) | India | South Africa |
Narrative | Physician certified CoD category |
---|---|
Heart failure. The patient death due to breathlessness. The person suffering paralysis and stroke lost on year with chest pain very pressure after then person was head. | Cardiovascular disease |
One day 13/03/01 he fell ill with some fever and chest pain who called the Doctor. On 15/03/01 the deceased was crying in the chest pain and high fever. We were ready to shift. The patient to the Hospital, some water came out from the deceased mouth and closed his eyes and passed away. | Acute respiratory infections |
Implementation of metrics
Machine learning models for text classification
Results
Precision | Sensitivity | F 1 | PCCC | CSMFA | CCCSMFA | |
---|---|---|---|---|---|---|
Adult (15–69 years) | ||||||
Naïve Bayes | .710 | .710 | .704 | .689 | .929 | .801 |
Random forest | .733 | .730 | .728 | .711 | .948 | .854 |
SVM | .746 | .737 | .740 | .718 |
.962
|
.894
|
Neural network |
.773
|
.770
|
.770
|
.764
|
.962
|
.894
|
Child (29 days–14 years) | ||||||
Naïve Bayes | .647 | .595 | .608 | .565 | .851 | .585 |
Random forest | .687 | .620 | .638 | .591 | .872 | .643 |
SVM | .686 | .658 | .666 | .632 |
.914
|
.760
|
Neural network |
.719
|
.695
|
.698
|
.672
| .904 | .733 |
Neonate (<29 days) | ||||||
Naïve Bayes | .507 | .516 | .493 | .376 | .826 | .509 |
Random forest | .534 | .542 | .524 | .411 | .852 | .581 |
SVM | .537 | .538 | .524 | .404 |
.857
|
.597
|
Neural network |
.579
|
.576
|
.556
|
.453
| .825 | .507 |
Precision | Sensitivity | F1 | PCCC | CSMFA | CCCSMFA | |
---|---|---|---|---|---|---|
Adult (15–69 years) | ||||||
Naïve Bayes | .591 | .593 | .580 | .583 | .869 | .643 |
Random forest | .644 | .647 | .634 | .638 | .905 | .742 |
SVM |
.665
|
.662
|
.655
|
.654
|
.908
|
.751
|
Neural network | .630 | .654 | .620 | .646 | .840 | .567 |
Child (29 days–14 years) | ||||||
Naïve Bayes | .493 | .402 | .427 | .379 | .768 | .369 |
Random forest |
.570
| .507 | .514 | .488 |
.807
|
.476
|
SVM | .567 |
.530
|
.528
|
.512
| .796 | .446 |
Neural network | .512 | .494 | .474 | .474 | .753 | .330 |
Neonate (<29 days) | ||||||
Naïve Bayes | .434 | .469 | .435 | .399 | .797 | .448 |
Random forest | .424 | .455 | .426 | .384 | .798 | .450 |
SVM |
.505
|
.497
|
.476
|
.431
|
.813
|
.492
|
Neural network | .328 | .361 | .306 | .278 | .634 | .007 |
Precision | Sensitivity | F 1 | PCCC | CSMFA | CCCSMFA | |
---|---|---|---|---|---|---|
Adult (15–69 years) | ||||||
Naïve Bayes | .517 | .517 | .513 | .481 |
.932
|
.814
|
Random forest | .511 | .517 | .496 | .480 | .844 | .577 |
SVM | .569 | .566 | .561 | .543 | .901 | .730 |
Neural network |
.575
|
.578
|
.570
|
.547
| .918 | .777 |
Child (29 days–14 years) | ||||||
Naïve Bayes | .488 | .440 | .435 | .395 | .761 | .351 |
Random forest | .521 | .502 | .487 | .463 | .816 | .501 |
SVM | .535 | .518 | .512 | .479 |
.872
|
.653
|
Neural network |
.572
|
.562
|
.552
|
.527
| .869 | .645 |
Neonate (<29 days) | ||||||
Naïve Bayes |
.532
|
.526
|
.483
|
.404
| .702 | .191 |
Random forest | .409 | .496 | .427 | .366 |
.710
|
.213
|
SVM | .387 | .417 | .371 | .266 | .693 | .165 |
Neural network | .356 | .412 | .354 | .259 | .636 | .012 |
Precision | Sensitivity | F 1 | PCCC | CSMFA | CCCSMFA | |
---|---|---|---|---|---|---|
Adult (15–69 years) | ||||||
Naïve Bayes | .433 | .448 | .431 | .432 |
.876
|
.662
|
Random forest | .438 | .464 | .436 | .448 | .832 | .543 |
SVM |
.502
|
.505
|
.491
| .490 | .857 | .612 |
Neural network | .470 | .495 | .451 | .480 | .750 | .322 |
Child (29 days–14 years) | ||||||
Naïve Bayes | .378 | .388 | .370 | .360 | .793 | .437 |
Random forest | .456 | .450 | .431 | .425 | .799 | .453 |
SVM |
.471
|
.465
|
.452
|
.440
|
.816
|
.499
|
Neural network | .388 | .428 | .374 | .402 | .667 | .095 |
Neonate (<29 days) | ||||||
Naïve Bayes | .276 | .384 | .305 | .296 | .610 | -.060 |
Random forest | .292 | .369 | .314 | .279 | .673 | .111 |
SVM |
.391
|
.405
|
.373
|
.320
|
.733
|
.274
|
Neural network | .156 | .265 | .179 | .160 | .502 | -.353 |