Reading Data
This section demonstrates how the isx package can be used to read data from Inscopix files.
Refer to the file types table for reference on the Inscopix file types and read support.
Note
The following sections assume the isx package has been imported, i.e., import isx
Microscope & Behavior Movies
Microscope Movie and Behavior Movie file types can be read using the isx.Movie class.
# Open the movie for reading
movie = isx.Movie.read("movie.isxd")
Movie objects have timing and spacing properties which can be accessed:
# Timing information can be accessed using:
movie.timing
# The number of frames can be accessed using:
movie.timing.num_samples
# The time period can be accessed using:
movie.timing.period.secs_float
# Spacing information can be accessed using:
movie.spacing
# The dimensions of the frame can be accessed using:
# which will return a 2-tuple containing the dimensions of the frame
# (num_rows, num_cols), or (height, width)
movie.spacing.num_pixels
Frames from movies can be read into memory:
# The pixel data type can accessed using:
movie.data_type
# Read every frame of the movie, and process it
for i in range(movie.timing.num_samples):
# Dimensions of the frame are (num_rows, num_cols), or (height, width)
frame = movie.get_frame_data(i)
# Process frame
...
Warning
It’s recommended to read one frame into memory at a time to prevent out of memory errors.
Note
Python indexes by 0, so the first frame is at index 0, and the the second frame is at index 1, and so on.
Cell Sets
Cell Set file types can be read using the isx.CellSet class.
# Open the cell set for reading
cell_set = isx.CellSet.read("cell_set.isxd")
Similar to movies, cell sets have timing and spacing properties which can be accessed. These properties are derived from the parent movie which generated the cell set.
# Timing information can be accessed using:
cell_set.timing
# The number of frames can be accessed using:
cell_set.timing.num_samples
# The time period can be accessed using:
cell_set.timing.period.secs_float
# Spacing information can be accessed using:
cell_set.spacing
# The dimensions of the frame can be accessed using:
# which will return a 2-tuple containing the dimensions of the frame
# (num_rows, num_cols), or (height, width)
cell_set.spacing.num_pixels
Cell data from cell sets can be read into memory:
# The number of cells in the cell set can accessed using:
cell_set.num_cells
# Read data of every cell in the cell set, and process it
for i in range(cell_set.num_cells):
# Get the cell name
name = cell_set.get_cell_name(i)
# Get the temporal activity trace of a cell
trace = cell_set.get_cell_trace(i)
# Get the spatial footprint of a cell
footprint = cell_set.get_cell_image_data(i)
# Process cell data
...
Event Sets
Event Set file types can be read using the isx.EventSet class.
# open the event set for reading
event_set = isx.EventSet.read("event_set.isxd")
Similar to cell sets, event sets have timing properties which can be accessed. These properties are derived from the parent cell set which generated the event set.
# Timing information can be accessed using:
event_set.timing
# The number of frames can be accessed using:
event_set.timing.num_samples
# The time period can be accessed using:
event_set.timing.period.secs_float
Cell data from event sets can be read into memory:
# Read data of every cell in the event set, and process it
for i in range(event_set.timing.num_samples):
# Get the cell name
name = event_set.get_cell_name(i)
# Get the event timestamps and amplitudes of a cell
offsets, amplitudes = event_set.get_cell_data(i)
# Process cell data
...
Vessel Sets
Vessel Set file types can be read using the isx.VesselSet class.
# Open the vessel set for reading
vessel_set = isx.VesselSet.read("vessel_set.isxd")
Similar to movies, vessel sets have timing and spacing properties which can be accessed. These properties are derived from the parent movie which generated the vessel set.
# Timing information can be accessed using:
vessel_set.timing
# The number of frames can be accessed using:
vessel_set.timing.num_samples
# The time period can be accessed using:
vessel_set.timing.period.secs_float
# Spacing information can be accessed using:
vessel_set.spacing
# The dimensions of the frame can be accessed using:
# which will return a 2-tuple containing the dimensions of the frame
# (num_rows, num_cols), or (height, width)
vessel_set.spacing.num_pixels
Vessel data from vessel sets can be read into memory:
# The number of vessels in the vessel set can be accessed using:
vessel_set.num_vessels
# The standard deviation projection image of the parent movie
# This is the same for every vessel
image = vessel_set.get_vessel_image_data(0)
# Read data of every vessel in the vessel set, and process it
for i in range(vessel_set.num_vessels):
# Get the vessel name
name = vessel_set.get_vessel_name(i)
# Get the vessel status
status = vessel_set.get_vessel_status(i)
# For vessel diameter vessel sets, get the following data:
# The vessel diameter trace
trace = vessel_set.get_vessel_trace_data(i)
# The direction of rbc velocity trace
center_trace = vessel_set.get_vessel_center_trace_data(i)
# The vessel line points
line = vessel_set.get_vessel_line_data(i)
# For rbc velocity vessel sets, get the following data:
# The rbc velocity trace
trace = vessel_set.get_vessel_trace_data(i)
# The direction of rbc velocity trace
direction_trace = vessel_set.get_vessel_direction_trace_data(i)
# The first correlation heatmap of the vessel
corr = vessel_set.get_vessel_correlations_data(i, 0)
# The vessel box points
box = vessel_set.get_vessel_line_data(i)
# Process vessel data
...
GPIO & IMU
GPIO and IMU file types can be read using the isx.GpioSet class.
# open the gpio set for reading
gpio_set = isx.GpioSet.read("signals.gpio")
Gpio sets have timing properties which can be accessed.
# Timing information can be accessed using:
gpio_set.timing
# The number of frames can be accessed using:
gpio_set.timing.num_samples
# The time period can be accessed using:
gpio_set.timing.period.secs_float
Signal data from gpio sets can be read into memory:
# Number of channels (i.e., signals) can be accessed using:
gpio_set.num_channels
# Read data of every cell in the cell set, and process it
for i in range(gpio_set.num_channels):
# Get the channel name
name = gpio_set.get_channel_name(i)
# Get the signal timestamps and amplitudes of a channel
offsets, amplitudes = gpio_set.get_channel_data(i)
# Process signal data
...