diff --git a/bob/db/meds/database.py b/bob/db/meds/database.py
index 44ce728ef5e0c5396256e23bde4b5fda007229dd..b976e637695278d9348693a8c8189f291f861e78 100644
--- a/bob/db/meds/database.py
+++ b/bob/db/meds/database.py
@@ -68,17 +68,7 @@ class Database:
         )
 
         logger.debug(f"Reading '{subject_metadata_file}'.")
-        full_dataframe = pandas.read_csv(subject_metadata_file)
-        # Keep only these columns:
-        self.dataframe = full_dataframe[[
-            "img_dname",        # Directory name of the image
-            "img_name",         # Image file name
-            "subject_id",       # Unique int identifier for each subject
-            "subject_imgcount", # Image numerotation for each subject
-            "sex",              # Gender of the subject
-            "rac",              # Ethnicity of the subject
-            "label",            # Pose label (Frontal, Profile, ...)
-        ]]
+        self.dataframe = pandas.read_csv(subject_metadata_file)
         
         logger.debug(f"Filtering protocol genders.")
         self.dataframe = self.dataframe[
@@ -248,8 +238,8 @@ class Database:
 
         return retval
 
-    def _annotations(self, image_name):
-        """Returns the coordinates of both eyes in the image of a sample
+    def _sample_annotations(self, image_name):
+        """Returns the coordinates of both eyes in the image of a sample.
 
         Note:
             eyes positions correspond to the point of view of the subject:
@@ -270,11 +260,31 @@ class Database:
                 { "leye":(y_pos,x_pos), "reye":(y_pos,x_pos) }.
         """
         row = self.eyes_dataframe[self.eyes_dataframe.img_fname == image_name]
-        eyes = {
+        annotations = {
             "leye": (row.left_pupil_y.iloc[0], row.left_pupil_x.iloc[0]),
             "reye": (row.right_pupil_y.iloc[0], row.right_pupil_x.iloc[0])
         }
-        return eyes
+        return annotations
+
+    def _sample_gender(self, filename):
+        """Returns the gender of a given sample's subject."""
+        return self.dataframe[self.dataframe.img_name == filename].sex.iloc[0]
+
+    def _sample_ethnicity(self, filename):
+        """Returns the ethnicity of a given sample's subject."""
+        return self.dataframe[self.dataframe.img_name == filename].rac.iloc[0]
+
+    def _sample_height(self, filename):
+        """Returns the height of a subject in a given sample."""
+        return self.dataframe[self.dataframe.img_name == filename].hgt.iloc[0]
+
+    def _sample_weight(self, filename):
+        """Returns the weight of a subject in a given sample."""
+        return self.dataframe[self.dataframe.img_name == filename].wgt.iloc[0]
+
+    def _sample_pose(self, filename):
+        """Returns the facial pose of a subject in a sample."""
+        return self.dataframe[self.dataframe.img_name == filename].label.iloc[0]
 
     def _subject_to_key(self, subject_id):
         """Transforms a subject ID to a unique int key
@@ -357,7 +367,12 @@ class Database:
                     ),
                     key=path,
                     subject=subject,
-                    annotations=self._annotations(filename),
+                    annotations=self._sample_annotations(filename),
+                    gender=self._sample_gender(filename),
+                    ethnicity=self._sample_ethnicity(filename),
+                    height=self._sample_height(filename),
+                    weight=self._sample_weight(filename),
+                    pose=self._sample_pose(filename),
                 )
             )
         return list(sets.values())