Spectroscopic data of pleural effusion

Description

Pleural effusion is a medical condition having its etiology in various primary diseases, eg. cardiological, pulmonological, and oncological. It is a side effect of diseases such as pneumonia, tuberculosis, cancer, heart failure, and inflammation which are responsible for causing inflammatory diseases of the pleura. Late detection of pleural effusion disturbs the respiratory function, heart operation or may cause bleeding into the pleural cavity. In such a case, there is an immediate requirement for medical intervention and drainage of the pleural space or execution of pleurodesis causing pleural plaques to fuse, which is an invasive and potentially life-threatening procedure. Therefore, patients with fluid in the pleural cavity, especially a group with recurrent pleural effusion should be continuously monitored. In a standard clinical procedure, the presence of pleural effusion is diagnosed by a physician with the use of chest radiography, ultrasonography, or computed tomography. Although ultrasound becomes portable and handheld the diagnosis is possible only in the clinics or medical centers so far. The monitoring of pleural effusion in outpatients is still an unsolved problem. So far, the use of near-infrared spectroscopy (NIRS) to test the amount of hydro-derivative fluids in various biomedical applications has been developed. For example, a method for assessing brain edema, water composition in rabbit meal, the water content in moisturizing creams or ceramic plasters has been published. In this work, the data form noninvasive NIRS measurement in patients with pleural effusion were collected. The method for noninvasive detection of pleural effusion using continuous-wave near-infrared spectroscopy is presented. The method has been developed based on numerical simulations of light transport in turbid media and validated on phantoms mimicking tissues and on patients with excess pleural fluid undergoing thoracocentesis. Numerical simulations consisted in modeling light propagation in an optically turbid, non-homogeneous medium with tissue properties of peripulmonary structures and for various amounts of pleural fluid. A three-layer model depicting pleural effusion was adopted. It was assumed that the outer surface layer consists of skin, fat and muscle. Fiber optics were connected to this layer during measurements. Two fiber optics were used: one guiding the light from the source (emitter) and the second delivering to the spectrometer (detector). The second layer, present only when pleural effusion is present, is water layer. The third layer is the lung tissue layer. The NIRS measurement was performed in a reflective scenario.

Steps to reproduce

The measurements were done on patients with pleural fluid, before and after decompression. The measurement results were verified using an ultrasound method, which determined the area of fluid occurrence and its depth from the body surface (pleural fluid depth). Spectrometric measurements were performed in two wavelength ranges: 1. Hammamatsu C10083CA, spectral response range: 320-1000nm, spectral resolution 8nm (LowNIR). The distance between the emitter and the detector was 55mm; 2. Hammatsu C9406GC, spectral response range 900-1700nm, spectral resolution 7nm (HighNIR). The distance between the emitter and the detector was 20mm The maximum distances at which reliable detection of the optical signal were possible were used. A transmitting optical fiber bundle with a diameter of 4mm and a length of 1.5m from Lupek was used, as well as a receiving optical fiber from Thorlabs M35L01 with a length of 1m. An infrared lamp R123RB E27 was used as the light source. Measurement procedure During the measurement, the patient remained in a sitting position. First, measurements were made in the LowNIR band, then in the HighNIR band. Before decompression 1. The area of pleural fluid occurrence was determined using an ultrasound method, this range was marked on the patient's body with a felt-tip pen 2. Measurements were made in the marked area of fluid, starting from the lowest level and then moving the measurement pad up by 2 cm for subsequent measurements. After decompression 1. The absence of fluid was checked using ultrasound 2. Measurements were made in the marked area of fluid presence before decompression, starting from the lowest level and then moving the measurement pad up 2 cm for subsequent measurements. Measurement parameters: Integration time (photon collection time individually for each wavelength) was set in the spectrometers to 100ms, the number of scans was 1000, averaged in the spectrometer. Hence, the time of a single measurement was 100 seconds. Data processing 1. The spectral characteristics were determined for each measurement. 2. Reference bands have been determined in which the signal variability for subsequent measurements is low. They were, respectively, for LowNIR: 959.75 – 1039.90 nm and for HighNIR 1404.40–1697.40nm. 3. Normalization of the spectral characteristics was performed. 4. Averaging of the measured signals was performed, independently in the LowNIR and HighNIR bands, as well as measurements before and after decompression. Four averaged spectral characteristics were obtained for a single patient. 5. Independently for the LowNIR and HighNIR bands, the differential characteristics were calculated by subtracting the spectral characteristics: after decompression LowNIR minus before decompression LowNIR and after decompression HighNIR minus before decompression HighNIR. In this way, differential spectral characteristics were obtained independently for the LowNIR and HighNIR bands.

Categories

Funding

Licence