Update NeuroScope2.m with power calculation #134

MingzeDou · 2025-12-18T13:05:50Z

No description provided.

Copilot

Pull request overview

This PR adds a new power calculation feature to NeuroScope2, allowing users to compute and visualize spectral power across channels for a specified time window and frequency band. The implementation includes interactive dual-plot visualization (spatial map and channel-ordered line plot) with click-to-identify functionality.

Key Changes:

Added "Power Calculation" menu item under the Analysis menu
Implemented calculatePower function with frequency band filtering and power computation
Added powerPlotClickCallback for interactive plot navigation
Updated UI panel minimum height to accommodate new functionality
Various minor formatting fixes (trailing whitespace removal)

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2025-12-18T13:16:45Z

NeuroScope2.m

+        try
+            % fseek to position
+            % Position in bytes: sample_index * nChannels * bytes_per_sample (2 for int16)
+            offset_bytes = start_sample * nChannels * 2;


The hardcoded value '2' for bytes per sample assumes int16 precision. However, the precision is read from UI.settings.precision which could potentially be different (e.g., 'int32', 'single', 'double'). This could cause incorrect file positioning and data corruption if the actual data type differs from int16.

Copilot · 2025-12-18T13:16:45Z

NeuroScope2.m

+        if isnan(t_start) || isnan(t_end) || t_end <= t_start
+            MsgLog('Invalid Time Range.', 4);
+            return;
+        end
+
+        if isnan(low_freq) || isnan(high_freq) || high_freq <= low_freq
+            MsgLog('Invalid Frequency Range.', 4);
+            return;
+        end
+
+        duration = t_end - t_start;


No validation that t_start is non-negative or that t_end doesn't exceed the recording duration. Users could request data beyond file boundaries, causing fseek/fread errors. Add bounds checking against the total recording time available.

Copilot · 2025-12-18T13:16:45Z

NeuroScope2.m

+
+        % Hidden/Unused controls for compatibility if needed, or remove.
+        % Keeping them valid but hidden is safer if logic depends on them, but we hardcode 4-12Hz now.
+        % So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will removing them from UI creation.


The comment states "I will removing them from UI creation" which is grammatically incorrect and unclear. Should be "I will remove them from UI creation" or "I am removing them from UI creation".

Suggested change

% So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will removing them from UI creation.

% So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will remove them from UI creation.

Copilot · 2025-12-18T13:16:46Z

NeuroScope2.m

+        try
+            % fseek to position
+            % Position in bytes: sample_index * nChannels * bytes_per_sample (2 for int16)
+            offset_bytes = start_sample * nChannels * 2;
+            fseek(curr_fid, offset_bytes, 'bof');
+
+            % fread
+            raw_data = fread(curr_fid, [nChannels, n_samples], precision);
+
+            % Transpose to (samples x channels) and scale
+            traces = double(raw_data') * UI.settings.leastSignificantBit;


Loading potentially large amounts of data (duration * sampling rate * channels) into memory without size validation. For long durations or high sampling rates, this could cause memory exhaustion. Consider adding a check for maximum data size or implementing chunked processing for large requests.

Copilot · 2025-12-18T13:16:46Z

NeuroScope2.m

+            [b, a] = butter(2, [low_freq, high_freq] / nyquist, 'bandpass');
+        end
+
+        filtered_traces = filtfilt(b, a, traces);


The filtfilt operation is applied to all channels simultaneously without try-catch protection. If the filter becomes unstable during execution (despite the stability check), this will crash the function. Wrap filtfilt in error handling to provide graceful failure.

Suggested change

filtered_traces = filtfilt(b, a, traces);

try

filtered_traces = filtfilt(b, a, traces);

catch ME

MsgLog(sprintf('Bandpass filtering failed (%s). Using unfiltered traces instead.', ME.message), 3);

filtered_traces = traces;

end

Copilot · 2025-12-18T13:16:46Z

NeuroScope2.m

        % Defining flexible panel heights
        set(UI.panel.analysis.main, 'Heights', [150 60 100 100 100],'MinimumHeights',[150 60 100 100 100]);
        UI.panel.analysis.main1.MinimumWidths = 218;
-        UI.panel.analysis.main1.MinimumHeights = 510;
+        UI.panel.analysis.main1.MinimumHeights = 710;


The MinimumHeights value changed from 510 to 710, presumably to accommodate the new Power Calculation menu item. However, this is a significant UI change (200 pixel increase) that may affect the layout on smaller screens or cause UI overlap. Verify that this works correctly across different screen resolutions and consider making this more flexible.

Suggested change

UI.panel.analysis.main1.MinimumHeights = 710;

% Set analysis panel minimum height adaptively to work across screen sizes

screenSize = get(0,'ScreenSize'); % [left bottom width height]

screenHeight = screenSize(4);

desiredMinHeight = 710; % height to accommodate new menu item

baseMinHeight = 510; % previous minimum height (fallback)

maxAllowedHeight = max(baseMinHeight, screenHeight - 150); % keep some vertical margin

minAnalysisHeight = min(desiredMinHeight, maxAllowedHeight);

UI.panel.analysis.main1.MinimumHeights = minAnalysisHeight;

Copilot · 2025-12-18T13:16:47Z

NeuroScope2.m

+        end
+
+        filtered_traces = filtfilt(b, a, traces);
+


Filter stability check using roots(a) may fail for edge cases or produce numerical instability warnings. After reducing to order 2, there's no verification that the second-order filter is stable, and no fallback if it also fails. Consider adding a stability check after order reduction or using filtfilt's error handling.

Suggested change

end

filtered_traces = filtfilt(b, a, traces);

% Re-check stability after order reduction

if any(abs(roots(a)) >= 1)

MsgLog('Second-order filter also unstable. Skipping filtering and using raw traces.', 4);

filtered_traces = traces;

else

try

filtered_traces = filtfilt(b, a, traces);

catch ME

MsgLog(['Filtering failed for second-order filter: ' ME.message], 4);

filtered_traces = traces;

end

end

else

% Original 3rd-order filter is stable

try

filtered_traces = filtfilt(b, a, traces);

catch ME

MsgLog(['Filtering failed for third-order filter: ' ME.message], 4);

filtered_traces = traces;

end

end

Copilot · 2025-12-18T13:16:47Z

NeuroScope2.m

+
+        % Bandpass Filter
+        % Downgraded to 3rd order Butterworth to improve stability for narrow/low bands
+        nyquist = fs / 2;


The filter normalization check (high_freq / nyquist) should ensure that the normalized frequencies are less than 1.0 to avoid filter design errors. Add validation to check that high_freq is less than the Nyquist frequency (fs/2) before filter design.

Suggested change

nyquist = fs / 2;

nyquist = fs / 2;

% Ensure band limits are valid relative to Nyquist to avoid filter design errors

if high_freq >= nyquist

adjHigh = nyquist - eps(nyquist);

MsgLog(sprintf('High cutoff frequency (%.4f Hz) >= Nyquist (%.4f Hz). Adjusting to %.4f Hz.', ...

high_freq, nyquist, adjHigh), 3);

high_freq = adjHigh;

end

if low_freq < 0

MsgLog(sprintf('Low cutoff frequency (%.4f Hz) < 0 Hz. Adjusting to 0 Hz.', low_freq), 3);

low_freq = 0;

end

if low_freq >= high_freq

% Ensure a valid band; nudge low_freq below high_freq if needed

newLow = max(0, high_freq * 0.5);

MsgLog(sprintf('Low cutoff frequency (%.4f Hz) >= high cutoff (%.4f Hz). Adjusting low cutoff to %.4f Hz.', ...

low_freq, high_freq, newLow), 3);

low_freq = newLow;

end

Copilot · 2025-12-18T13:16:47Z

NeuroScope2.m

+        % 4-12Hz Power Heatmap
+        % Power Calculation
+
+        % Hidden/Unused controls for compatibility if needed, or remove.
+        % Keeping them valid but hidden is safer if logic depends on them, but we hardcode 4-12Hz now.
+        % So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will removing them from UI creation.
+


Incomplete or obsolete comments that don't explain the purpose. These comments mention "4-12Hz Power Heatmap" and "Hidden/Unused controls" but don't provide clear context about what was removed or why. Either complete the explanation or remove these placeholder comments.

Copilot · 2025-12-18T13:16:48Z

NeuroScope2.m

+            set(plotData.hInfo, 'String', txt_str);
+        end
+    end
+
    function setSortingMetric(~,~)


The setSortingMetric function body was removed, leaving only the uiresume call. If the function previously set UI.params.sortingMetric (as suggested by the function name), this functionality is now lost. Verify that removing this assignment doesn't break sorting metric selection functionality.

Suggested change

function setSortingMetric(~,~)

function setSortingMetric(~,~)

UI.params.sortingMetric = UI.panel.cell_metrics.sortingMetric.String{UI.panel.cell_metrics.sortingMetric.Value};

Copilot AI review requested due to automatic review settings December 18, 2025 13:05

Copilot started reviewing on behalf of MingzeDou December 18, 2025 13:06 View session

Update NeuroScope2.m

d1ebc61

Copilot AI reviewed Dec 18, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Update NeuroScope2.m with power calculation #134

Update NeuroScope2.m with power calculation #134

Uh oh!

MingzeDou commented Dec 18, 2025

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Copilot AI Dec 18, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

	% So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will removing them from UI creation.
	% So strictly speaking we don't need them in UI, but to avoid errors if I miss any reference, I will remove them from UI creation.

-        filtered_traces = filtfilt(b, a, traces);
+        try
+            filtered_traces = filtfilt(b, a, traces);
+        catch ME
+            MsgLog(sprintf('Bandpass filtering failed (%s). Using unfiltered traces instead.', ME.message), 3);
+            filtered_traces = traces;
+        end

-        UI.panel.analysis.main1.MinimumHeights = 710;
+        % Set analysis panel minimum height adaptively to work across screen sizes
+        screenSize = get(0,'ScreenSize');              % [left bottom width height]
+        screenHeight = screenSize(4);
+        desiredMinHeight = 710;                        % height to accommodate new menu item
+        baseMinHeight = 510;                           % previous minimum height (fallback)
+        maxAllowedHeight = max(baseMinHeight, screenHeight - 150); % keep some vertical margin
+        minAnalysisHeight = min(desiredMinHeight, maxAllowedHeight);
+        UI.panel.analysis.main1.MinimumHeights = minAnalysisHeight;

-        nyquist = fs / 2;
+        nyquist = fs / 2;
+        % Ensure band limits are valid relative to Nyquist to avoid filter design errors
+        if high_freq >= nyquist
+            adjHigh = nyquist - eps(nyquist);
+            MsgLog(sprintf('High cutoff frequency (%.4f Hz) >= Nyquist (%.4f Hz). Adjusting to %.4f Hz.', ...
+                           high_freq, nyquist, adjHigh), 3);
+            high_freq = adjHigh;
+        end
+        if low_freq < 0
+            MsgLog(sprintf('Low cutoff frequency (%.4f Hz) < 0 Hz. Adjusting to 0 Hz.', low_freq), 3);
+            low_freq = 0;
+        end
+        if low_freq >= high_freq
+            % Ensure a valid band; nudge low_freq below high_freq if needed
+            newLow = max(0, high_freq * 0.5);
+            MsgLog(sprintf('Low cutoff frequency (%.4f Hz) >= high cutoff (%.4f Hz). Adjusting low cutoff to %.4f Hz.', ...
+                           low_freq, high_freq, newLow), 3);
+            low_freq = newLow;
+        end

	function setSortingMetric(~,~)
	function setSortingMetric(~,~)
	UI.params.sortingMetric = UI.panel.cell_metrics.sortingMetric.String{UI.panel.cell_metrics.sortingMetric.Value};

Update NeuroScope2.m with power calculation #134

Are you sure you want to change the base?

Update NeuroScope2.m with power calculation #134

Uh oh!

Conversation

MingzeDou commented Dec 18, 2025

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Copilot AI Dec 18, 2025

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant