Multidimensional Dataset for APA Investigations in Cancer Patients
DOI10.5281/zenodo.13711426Zenodo13711426MaRDI QIDQ6718759FDOQ6718759
Dataset published at Zenodo repository.
F. Amato, Vittorio Santoriello, Alfonso Maria Ponsiglione, Maria Romano, Marco Cascella, Ornella Piazza
Publication date: 6 September 2024
Copyright license: Creative Commons Attribution 4.0 International
The dataset contains data collected from patients suffering from cancer-related pain. The features extracted from clinical data (including typical cancer phenomena such as breakthrough pain) and the biosignal acquisitions contributed to the definition of a multidimensional dataset. This unique database can be useful for the characterization of the patients pain experience from a qualitative and quantitative perspective. We implemented measurable biosignals-related indicators of the individuals pain response and of the overall Autonomic Nervous System (ANS) functioning. The most peculiar features extracted from EDA and ECG signals can be adopted to investigate the status andcomplex functioning of the ANS through the study of sympatho-vagal activations. Specifically, while EDA is mainly related sympathetic activation, the Heart Rate Variability (HRV), which can be derived from ECG recordings, is strictly related to the interplay between sympathetic and parasympathetic functioning. As far as the EDA signal, two types of analyzes have been performed: (i) the Trough-To-Peak analysis (TTP), or min-max analysis, aimed at measuring the difference between the Skin Conductance (SC) at the peak of a response and its previous minimum within pre-established time-windows; (ii) the Continuous Decomposition Analysis (CDA), aimed at performing a decomposition of SC data into continuous signals of tonic (basic level of conductance) and phasic (short-duration changes in the SC) activity. Before applying the TPP analysis or the CDA, the signal was filtered by means of a fifth-order Butterworth low-pass filter with a cutoff frequency of 1 Hz and downsampled up to 10 Hz to reducing the computational burden of the analysis. The application of TPP and CDA allowed the detection and measurement of SC Responses (SCR) and the following parameters have been calculated for both TPP and CDA methodologies: Total number of detected SCRs. Maximum value of SCRs [measured in S]. Minimum value of SCRs [measured in S]. Arithmetic mean of the SCRs [measured in S]. Maximum interval between SCRs [measured in ms]. Minimum interval between SCRs [measured in ms]. Arithmetic mean of the intervals between SCRs [measured in ms]. Concerning the ECG, the RR series of interbeat intervals (i.e., the time between successive R waves of the QRS complex on the ECG waveform) has been computed to extract time-domain parameters of the HRV. The R peak detection was carried out by adopting the PanTompkins algorithm for QRS detection and R peak identification. The corresponding RR series of interbeat intervals were derived as the difference betweensuccessive R peaks. The ECG-derived RR time series was then filtered by means of a recursive procedure to remove the intervals differing most from the mean of the surrounding RR intervals. Then, both the Time-Domain Analysis (TDA) and Frequency-Domain Analysis (FDA) of the HRV have been carried out to extract the main features characterizing the variability of the heart rhythm. Time-domain parameters are obtained from statistical analysis of the intervals between heart beats and are used to describe how much variability in the heartbeats is present at various time scales. The parameters computed through the TDA include the following: Arithmetic mean of the RR time series [measured in ms]. The standard deviation of the RR time series [measured in ms]. Mean value of heart rate [measured in bpm]. Standard deviation of the heart rate [measured in bpm]. Root Mean Square of Successive Differences of RR intervals [measured in ms], which is sensitive to high-frequency heart period fluctuations in the respiratory frequency range and has been used asan index of vagal cardiac control. Number of successive RR intervals whose difference is higher than 50 ms. Percentage of successive RR intervals higher than 50 ms. Number of successive RRintervals whose difference is higher than 50 ms. Frequency-domain parameters reflect the distribution of spectral power across different frequencies bands and are used to assess specific components of HRV (e.g., thermoregulation control loop, baroreflex control loop, and respiration control loop, which are regulated by both sympathetic and vagal nerves of the ANS).The parameters computed through the FDA have been computed by adopting the Welch's Fourier periodogram method based on the Discrete Fourier Transform (DFT), which allows the expression of the RR series in the discrete frequency domain. However, due to the non-stationarity of the RR series, Welch Fourier periodogram method is used for dealing with non-stationarity. Specifically, Welch's periodogram divides the signal into specific periods of constant length appliying the Fast Fourier Transform (FFT) trasforming individually these parts of the signal. The periodogram is basically a way of estimating power spectral density of a time series. The FDA parameters include the following: Peak value in the Very Low Frequency Band of the HRV power density spectrum [measured in Hz]. Peak value in the Low Frequency Band of the HRV power density spectrum [measured in Hz]. Peak value in the High Frequency Band of the HRV power density spectrum [measured in Hz]. Power in the Very Low Frequency Band of the HRV power density spectrum [measured in ms^2]. Power in the Low Frequency Band of the HRV power density spectrum [measured in ms^2]. Power in the High Frequency Band of the HRTotal Power of the HRV power density spectrum [measured in ms^2]. Total Power of the HRV power density spectrum [measured in ms^2]. Percentage power in the Very Low Frequency Band of the HRV power density spectrum with respect to the total power. Percentage power in the Low Frequency Band of the HRV power density spectrum with respect to the total power. Percentage power in the High Frequency Band of the HRV power density spectrum with respect to the total power. Normalized power in the Low Frequency Band of the HRV power density spectrum with respect to the sum of LF and HF power. Normalized power in the High Frequency Band of the HRV power density spectrum with respect to the sum of LF and HF power. Sympathovagal balance measured as the ration between power in LF and power in the LF band.
This page was built for dataset: Multidimensional Dataset for APA Investigations in Cancer Patients
