Remove dependency and add annotation functionality

pyproject.toml

@@ -19,7 +19,7 @@ packages = ["src/epilepsy2bids"]
 
 [project]
 name = "epilepsy2bids"
-version = "0.0.2"
+version = "0.0.3"
 authors = [
   { name="Jonathan Dan", email="[email protected]" },
   { name="PEDESITE" }
@@ -28,13 +28,13 @@ description = "Python library for converting EEG datasets of people with epileps
 readme = "README.md"
 requires-python = ">=3.10"
 dependencies = [
-  "nptyping>=2.5.0",
   "numpy>=1.24.3",
   "pandas>=2.1.3",
   "pyarrow>=16.1.0",
   "pyedflib>=0.1.36",
   "resampy>=0.4.2",
-  "termcolor>=2.4.0"
+  "termcolor>=2.4.0",
+  "timescoring>=v0.0.5"
   ]
 classifiers = [
     "Programming Language :: Python :: 3",

requirements.txt

@@ -1,4 +1,3 @@
-nptyping
 numpy
 pandas
 pyarrow

src/epilepsy2bids/annotations.py

@@ -2,10 +2,11 @@
 import json
 from datetime import datetime
 from importlib import resources as impresources
-from typing import TypedDict
+from typing import List, Tuple, TypedDict
 
 import numpy as np
 import pandas as pd
+from timescoring.annotations import Annotation as MaskEvents
 
 from . import bids
 
@@ -73,6 +74,36 @@ def loadTsv(cls, filename: str):
 
         return annotations
 
+    @classmethod
+    def loadMask(cls, mask, fs):
+        maskEvent = MaskEvents(mask, fs)
+        return cls.loadEvents(maskEvent.events, len(mask) / fs)
+
+    @classmethod
+    def loadEvents(cls, events: List[Tuple[float, float]], duration: float):
+        annotations = cls()
+        for event in events:
+            annotation = Annotation()
+            annotation["onset"] = event[0]
+            annotation["duration"] = event[1] - event[0]
+            annotation["eventType"] = SeizureType.sz
+            annotation["confidence"] = "n/a"
+            annotation["channels"] = "n/a"
+            annotation["dateTime"] = "n/a"
+            annotation["recordingDuration"] = duration
+            annotations.events.append(annotation)
+        if len(events) == 0:
+            annotation = Annotation()
+            annotation["onset"] = 0
+            annotation["duration"] = duration
+            annotation["eventType"] = EventType.bckg
+            annotation["confidence"] = "n/a"
+            annotation["channels"] = "n/a"
+            annotation["dateTime"] = "n/a"
+            annotation["recordingDuration"] = duration
+            annotations.events.append(annotation)
+        return annotations
+
     def getEvents(self) -> list[(float, float)]:
         events = list()
         for event in self.events:
@@ -110,7 +141,10 @@ def saveTsv(self, filename: str):
                     line += "\t"
                 else:
                     line += "{}\t".format(event["channels"])
-                line += "{}\t".format(event["dateTime"].strftime("%Y-%m-%d %H:%M:%S"))
+                if isinstance(event["dateTime"], datetime):
+                    line += "{}\t".format(event["dateTime"].strftime("%Y-%m-%d %H:%M:%S"))
+                else:
+                    line += "{}\t".format(event["dateTime"])
                 line += "{:.2f}".format(event["recordingDuration"])
                 line += "\n"
                 f.write(line)

src/epilepsy2bids/eeg.py

@@ -10,7 +10,6 @@
 import pandas as pd
 import pyedflib
 import resampy
-from nptyping import Float, NDArray, Shape
 
 
 class FileFormat(str, enum.Enum):
@@ -113,7 +112,7 @@ class FileHeader(TypedDict):
 
     def __init__(
         self,
-        data: NDArray[Shape["*, *"], Float],
+        data,
         channels: tuple[str],
         fs: int,
         montage: Montage = Montage.UNIPOLAR,
@@ -209,18 +208,18 @@ def loadEdfAutoDetectMontage(cls, edfFile: str):
         with pyedflib.EdfReader(edfFile) as edf:
             channel = edf.getLabel(0)
             edf._close()
-        if channel == Eeg.ELECTRODES_10_20[0]:
+        if channel == Eeg.ELECTRODES_10_20[0].upper():
             montage = Eeg.Montage.UNIPOLAR
             electrodes = Eeg.ELECTRODES_10_20
-        elif channel == Eeg.BIPOLAR_DBANANA[0]:
-            montage = Eeg.Montage.BIPOLAR_DBANANA
+        elif channel == Eeg.BIPOLAR_DBANANA[0].upper():
+            montage = Eeg.Montage.BIPOLAR
             electrodes = Eeg.BIPOLAR_DBANANA
         else:
             raise ValueError(
                 f"Unrecognized electrode: {channel}. Expected {Eeg.ELECTRODES_10_20[0]} or {Eeg.BIPOLAR_DBANANA[0]}"
             )
 
-        return Eeg.loadEdf(edfFile, montage, electrodes)
+        return cls.loadEdf(edfFile, montage, electrodes)
 
     def resample(self, newFs: int):
         """Resample data to a new sampling frequency.