cinc2023/utils/features.py at master · DeepPSP/cinc2023 · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
"""
features from metadata
"""

from typing import Union

import numpy as np
import pandas as pd

from helper_code import get_age, get_cpc, get_ohca, get_outcome, get_rosc, get_sex, get_shockable_rhythm, get_ttm  # get_vfib,

__all__ = [
    "get_features",
    "get_labels",
]


def get_features(patient_metadata: str, ret_type: str = "np") -> Union[np.ndarray, pd.DataFrame, dict]:
    """Extract features from the patient metadata.

    Adapted from the official repo.

    Parameters
    ----------
    patient_metadata : str
        The patient metadata.
    ret_type : {"np", "pd", "dict"}
        The return type, by default "np".

    Returns
    -------
    np.ndarray or pd.DataFrame or dict
        The patient features.

    """
    age = get_age(patient_metadata)
    sex = get_sex(patient_metadata)
    rosc = get_rosc(patient_metadata)
    ohca = get_ohca(patient_metadata)
    vfib = get_shockable_rhythm(patient_metadata)
    ttm = get_ttm(patient_metadata)

    # Use one-hot encoding for sex; add more variables
    sex_features = np.zeros(2, dtype=int)
    if sex == "Female":
        female = 1
        male = 0
        other = 0
    elif sex == "Male":
        female = 0
        male = 1
        other = 0
    else:
        female = 0
        male = 0
        other = 1

    # Combine the patient features.
    if ret_type == "np":
        patient_features = np.array([age, female, male, other, rosc, ohca, vfib, ttm])
    elif ret_type == "pd":
        patient_features = pd.DataFrame(
            {
                "age": age,
                "sex_female": female,
                "sex_male": male,
                "sex_other": other,
                "rosc": rosc,
                "ohca": ohca,
                "vfib": vfib,
                "ttm": ttm,
            },
            index=[0],
        )
    elif ret_type == "dict":
        patient_features = {
            "age": age,
            "sex_female": female,
            "sex_male": male,
            "sex_other": other,
            "rosc": rosc,
            "ohca": ohca,
            "vfib": vfib,
            "ttm": ttm,
        }

    return patient_features


def get_labels(patient_metadata: str, ret_type: str = "dict") -> Union[np.ndarray, pd.DataFrame, dict]:
    """Extract labels from the patient metadata.

    Adapted from the official repo.

    Parameters
    ----------
    patient_metadata : str
        The patient metadata.
    ret_type : {"np", "pd", "dict"}
        The return type, by default "dict".

    Returns
    -------
    dict or np.ndarray or pd.DataFrame
        The patient labels, including
        - "outcome" (int)
        - "cpc" (float)

    """
    labels = {}
    labels["outcome"] = get_outcome(patient_metadata)
    labels["cpc"] = get_cpc(patient_metadata)

    if ret_type == "dict":
        pass
    elif ret_type == "np":
        labels = np.array([labels["outcome"], labels["cpc"]])
    elif ret_type == "pd":
        labels = pd.DataFrame(labels, index=[0])

    return labels