Humach Health: Risk Screening
Handle missing clinical values, prioritise sensitivity and reason about calibration in a screening setting.
Download mini_health.csvUse Python, SQL, Excel or your preferred workflow unless a task specifies otherwise.
Project dataset
Download CSVInspect the CSV before you answer the tasks
Scroll horizontally or vertically to examine the exact miniature data used by the validator.
| patient_id | age | bmi | systolic_bp | smoker | risk |
|---|---|---|---|---|---|
| P001 | 29 | 21.3 | 114 | 0 | 0 |
| P002 | 36 | 22.6 | 123 | 0 | 0 |
| P003 | 43 | 23.9 | 132 | 0 | 0 |
| P004 | 50 | 25.2 | 141 | 0 | 0 |
| P005 | 57 | 26.5 | 150 | 1 | 1 |
| P006 | 64 | 27.8 | 159 | 0 | 1 |
| P007 | 71 | 29.1 | 113 | 0 | 1 |
| P008 | 23 | 30.4 | 122 | 0 | 0 |
| P009 | 30 | 31.7 | 131 | 0 | 0 |
| P010 | 37 | 21.0 | 140 | 1 | 0 |
| P011 | 44 | 149 | 0 | 0 | |
| P012 | 51 | 23.6 | 158 | 0 | 0 |
| P013 | 58 | 24.9 | 112 | 0 | 0 |
| P014 | 65 | 26.2 | 121 | 0 | 0 |
| P015 | 72 | 27.5 | 130 | 1 | 0 |
| P016 | 24 | 28.8 | 139 | 0 | 0 |
| P017 | 31 | 30.1 | 148 | 0 | 1 |
| P018 | 38 | 31.4 | 157 | 0 | 1 |
| P019 | 45 | 20.7 | 111 | 0 | 0 |
| P020 | 52 | 22.0 | 120 | 1 | 0 |
| P021 | 59 | 23.3 | 129 | 0 | 0 |
| P022 | 66 | 138 | 0 | 0 | |
| P023 | 73 | 25.9 | 147 | 0 | 1 |
| P024 | 25 | 27.2 | 156 | 0 | 0 |
| P025 | 32 | 28.5 | 110 | 1 | 0 |
| P026 | 39 | 29.8 | 119 | 0 | 0 |
| P027 | 46 | 31.1 | 128 | 0 | 0 |
| P028 | 53 | 20.4 | 137 | 0 | 0 |
| P029 | 60 | 21.7 | 146 | 0 | 1 |
| P030 | 67 | 23.0 | 155 | 1 | 1 |
| P031 | 74 | 24.3 | 109 | 0 | 0 |
| P032 | 26 | 25.6 | 118 | 0 | 0 |
| P033 | 33 | 127 | 0 | 0 | |
| P034 | 40 | 28.2 | 136 | 0 | 0 |
| P035 | 47 | 29.5 | 145 | 1 | 1 |
| P036 | 54 | 30.8 | 154 | 0 | 1 |
| P037 | 61 | 20.1 | 108 | 0 | 0 |
| P038 | 68 | 21.4 | 117 | 0 | 0 |
| P039 | 75 | 22.7 | 126 | 0 | 0 |
| P040 | 27 | 24.0 | 135 | 1 | 0 |
Project workspace
Analyse → validate → submit.
Numeric and decision tasks are auto-checked. Written reasoning is checked for completeness and key ideas, then included in your local submission report.
0/5 validated
How many patients have risk=1?
For a skewed numeric feature with outliers, which simple imputation baseline is often safer than the mean?
For initial screening where missed positives are costly, which metric is especially important?
Explain why fitting an imputer on all patients before cross-validation is unsafe.
Submit your project locally
HumachLearn stores no server-side data. Submission creates a portable JSON report containing your answers, validation state and timestamp.