Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.
{
  "sampled": 3000,
  "single_modality_counts": {
    "CT": 1028,
    "ULTRASOUND": 402,
    "ANGIOGRAPHY": 203,
    "MRI": 502,
    "XRAY": 820,
    "PET": 18
  },
  "multi_modality_rows": 0,
  "no_modality_match_rows": 27,
  "top_30_cuis": [
    [
      "C0040405",
      1028
    ],
    [
      "C1306645",
      820
    ],
    [
      "C0024485",
      502
    ],
    [
      "C0041618",
      402
    ],
    [
      "C1999039",
      358
    ],
    [
      "C0817096",
      266
    ],
    [
      "C0002978",
      203
    ],
    [
      "C1996865",
      177
    ],
    [
      "C0037303",
      162
    ],
    [
      "C0023216",
      94
    ],
    [
      "C0000726",
      81
    ],
    [
      "C0205129",
      81
    ],
    [
      "C0030797",
      72
    ],
    [
      "C1140618",
      65
    ],
    [
      "C0037949",
      55
    ],
    [
      "nan",
      27
    ],
    [
      "C0032743",
      18
    ],
    [
      "C0006141",
      3
    ]
  ],
  "projection": {
    "CT": 27341,
    "ULTRASOUND": 10692,
    "ANGIOGRAPHY": 5399,
    "MRI": 13351,
    "XRAY": 21809,
    "PET": 479
  }
}

written: /home/minif/projects/claimflow/backend/ml_training/data/cui_probe.json
verdict: VIABLE (projected per-class: CT=27341 MRI=13351 XRAY=21809; need >=5000 each)