diff --git a/code/hrv/hrv_analysis.ipynb b/code/hrv/hrv_analysis.ipynb
index caf824c..374f8e6 100644
--- a/code/hrv/hrv_analysis.ipynb
+++ b/code/hrv/hrv_analysis.ipynb
@@ -230,7 +230,7 @@
     "    - physio_type (str, optional): If 'bbsig_preproc=True', type of physiological signal used. Default is 'ecg', can also be 'ppg'.\n",
     "\n",
     "    Returns: \n",
-    "    - A dictionary with two keys: 'RRI' (array of RR intervals in ms); 'RRI_Time' (array of cumulative R-peaks timestamps in ms). \n",
+    "    - A dictionary with two keys: 'RRI' (array of RR intervals in ms); 'RRI_Time' (array of cumulative R-peaks timestamps in seconds). \n",
     "    \"\"\"\n",
     "\n",
     "    ############## Data loading: BBSIG preprocessing files ##############\n",
@@ -288,15 +288,18 @@
     "\n",
     "    # Convert RR intervals from seconds to milliseconds, if necessary\n",
     "    if rri_unit == 's':\n",
-    "        rr_s = rr_s*1000 \n",
-    "        \n",
+    "        rr_ms = rr_s * 1000\n",
+    "    else:\n",
+    "        rr_ms = rr_s\n",
+    "\n",
     "    # Convert data into appropriate formats for NeuroKit2\n",
-    "    peaks = np.cumsum(rr_s)     # define R-peaks as the cumulative sum of RR intervals\n",
+    "    # NeuroKit2 expects RRI in milliseconds and RRI_Time in seconds.\n",
+    "    peaks_s = np.cumsum(rr_ms) / 1000  # R-peak timestamps in seconds\n",
     "\n",
     "    # Store data for participant in a dictionary\n",
     "    rrs_dict = {\n",
-    "        'RRI': np.array(rr_s),  # RR intervals in miliseconds\n",
-    "        'RRI_Time': peaks}      # R-peaks timestamps in seconds   \n",
+    "        'RRI': np.array(rr_ms),  # RR intervals in milliseconds\n",
+    "        'RRI_Time': peaks_s}     # R-peaks timestamps in seconds\n",
     "               \n",
     "    return rrs_dict"
    ]
@@ -324,7 +327,7 @@
     "def hrv_window_crop(rrs_dict, window_start=0, window_end=None, window_length=None):\n",
     "    \"\"\" Crop RR interval time series to a specified time window for HRV analysis.\n",
     "    Parameters:\n",
-    "    - rrs_dict (dict): Dictionary storing RR intervals in ms R-peaks timestamps in ms, generated by load_rr_data().\n",
+    "    - rrs_dict (dict): Dictionary storing RR intervals in ms and R-peaks timestamps in seconds, generated by load_rr_data().\n",
     "    - window_start (float): Start time of the window to be cropped (in seconds).\n",
     "    - window_end (float, optional): End time of the window to be cropped (in seconds).\n",
     "    - window_length (float, optional): Length of the window to be cropped (in seconds).\n",
@@ -355,7 +358,7 @@
     "\n",
     "    # Selected the R-peaks comprised between window_start and window_end\n",
     "    rrs_dict['RRI_Time'] = rrs_dict['RRI_Time'].tolist()\n",
-    "    rrt_cropped = [x for x in rrs_dict['RRI_Time'] if x>=window_start*1000 and x<=window_end*1000]\n",
+    "    rrt_cropped = [x for x in rrs_dict['RRI_Time'] if x >= window_start and x <= window_end]\n",
     "\n",
     "    # Get indices of window_start and window_end and extract RR interval durations comprised in-between\n",
     "    window_start_idx = rrs_dict['RRI_Time'].index(rrt_cropped[0])\n",
@@ -457,7 +460,10 @@
     "    # Calculate Frequency-domain HRV metrics using NeuroKit2\n",
     "    # Using Welch method and interpolation at 4Hz by default\n",
     "    # If plot is needed, set show=True\n",
-    "    hrv_f = nk.hrv_frequency(rrs_dict, interpolation_rate=interpolation_freq, \n",
+    "    # ulf=(0, 0) disables the ULF band, which otherwise returns spurious values\n",
+    "    # for short (<5 min) recordings and dominates the PSD plot.\n",
+    "    hrv_f = nk.hrv_frequency(rrs_dict, interpolation_rate=interpolation_freq,\n",
+    "                             ulf=(0, 0),\n",
     "                             normalize=False, psd_method='welch', show=show_plots) \n",
     "    \n",
     "    # Extract and format metrics\n",
diff --git a/code/preprocessing/ecg_preproc.ipynb b/code/preprocessing/ecg_preproc.ipynb
index 0ef3cb8..f9cf898 100644
--- a/code/preprocessing/ecg_preproc.ipynb
+++ b/code/preprocessing/ecg_preproc.ipynb
@@ -929,7 +929,7 @@
         "\n",
         "    # Interpolate HR values using Systole heart_rate()\n",
         "    # Hot fix @April 3rd, 2025: removed sfreq=sfreq from arguments due to bug with Systole mis-calculating HR (until it will be fixed)\n",
-        "    hr_bpm, hr_time = heart_rate(x=rr_tmp, unit='bpm', kind=interpol_type, input_type='rr_s') \n",
+        "    hr_bpm, hr_time = heart_rate(x=rr_tmp, output_unit='bpm', kind=interpol_type, input_type='rr_s') \n",
         "\n",
         "    # Define directory for TSV file with interpolated HR values\n",
         "    hr_tsv_fname = f'{bids_base_fname}_hr-bpm-{rr_type}.tsv.gz'\n",