====== SSc-ILD ======

  * GAP score
  * [[https://doi.org/10.1093/rheumatology/keaf445 | ILD-RISC score]]
    * Outcome to predict: the presence of ILD on HRCT
    * score: a probability (0 to 1.0)
    * ILD-RISC score >= 0.3 had sensitivity 85.6% and specificity 53.6%
    * items:
      * <todo>age</todo>
      * <todo>ever had digital ulcers (OR 2.06)</todo>
      * <todo>DLCO</todo>
      * <todo>FVC</todo>
      * <todo>antibody (topo I: OR 2.38, PM/SCL: 2.56, RNAPIII: 1.41)</todo>

====== IIM-ILD ======

===== FLAIR model =====

  * [[https://doi.org/10.1016/j.chest.2020.04.057 | article]]
    * bottom line: MDA5 is bad, high inflammatory markers are bad, l

^item^categorical_value^score^
| ferritin | >= 636 ng/mL | TRUE: 2, FALSE: 0|
| LDH | >= 355 U/L | TRUE: 2, FALSE: 0|
| anti-MDA5 antibody | >= 636 ng/mL | negative, 0; +, 2; ++, 3; +++, 4|
| high-resolution CT imaging score (as described by https://doi.org/10.1164/rccm.2106157)| >= 133 | TRUE: 3, FALSE: 0 |

  * Scoring
    * low, 0 to 4
    * medium, 5 to 9
    * high, 10 to 13

===== CROSS model =====

^boolean parameter^ $\beta$ coefficient ^ score ^
| CRP abnormal | 0.754 | 1 |
| Anti-Ro52 positivity | 1.505 | 2 |
| male sex | 0.624 | 1 |
| disease duration <3 months | 1.291 | 2 |

^ risk strata ^ qualifying scores ^
| moderate | 0-2 |
| high | 3-4 |
| very high | 5-6 |

===== RRP model =====

^boolean parameter^score if TRUE^
| fever | 5|
| periungual erythema | 6|
| elevated CRP | 5 |
| anti-MDA5 antibody | 8 |
| anti-Ro-52 antibody | 6 |


^ risk strata ^ qualifying scores ^
| moderate | 0-9 |
| high | 10-19 |
| very high | 20-30 |