Priorto begins by installing the Patient Monitor at each general ward bed, configured with a 12inch touchscreen and fiveparameter capability (ECG, NIBP, SpO₂, temperature, respiratory rate). The monitor is wallmounted on a swing arm that allows 180° rotation and 50 cm extension. Each monitor is connected to the central nursing station via a wired Ethernet network (1 Gbps backbone). A 4hour backup battery maintains operation during power fluctuations.

The Patient Monitor’s ECG module processes 3lead, 5lead, or 12lead signals. For telemetry wards, the 5lead configuration is standard, providing leads I, II, III, aVR, aVL, aVF, and V. The arrhythmia analysis algorithm is trained on a library of 10,000 annotated ECGs, with a sensitivity of 95% for ventricular tachycardia and 98% for asystole. The NIBP module uses an oscillometric method with a measurement range of 30–280 mmHg; the default interval is set to 60 minutes for stable patients, adjustable from 15 to 240 minutes.

The SpO₂ sensor uses a reusable finger clip (adult) or a wraparound tape (neonatal). The sensor’s signal quality index (SQI) is displayed as a bar graph; a low SQI triggers an alert to check the sensor placement. The temperature probe is available in oral, axillary, or rectal configurations; the default is axillary with a 5minute equilibration time. All vital signs are transmitted to the central station every 15 seconds for stable parameters, and every 2 seconds for arrhythmia events.

Technical Note: The 12lead ECG module requires a dedicated patient cable with 10 leads; using a 5lead cable will not activate the 12lead algorithm. The NIBP cuff should be sized correctly (bladder width 40% of arm circumference); a miscuff will cause errors of ±10 mmHg.

Priorto supplies the Holter Monitor as a standalone device for outpatient arrhythmia evaluation. The Holter recorder is a 3lead (or 12lead) device weighing 50 g, with a 7day battery life and 32 GB internal storage. The recorder is configured to continuously record all beats, with an event button for the patient to mark symptomatic periods. After the recording period, the device is docked into a workstation that downloads and analyzes the data using a convolutional neural network (CNN) algorithm.

The Holter Monitor’s 3lead configuration (modified V5, V1, and aVF) is standard for detecting supraventricular and ventricular arrhythmias. The 12lead option is specified for patients with known coronary artery disease or when STsegment analysis is required. The electrodes are made of solid gel with a hypoallergenic adhesive, rated for 7day wear. The recorder’s event button is backlit; the patient is instructed to press it once when experiencing palpitations, dizziness, or chest pain.

The analysis workstation’s CNN algorithm is trained on 50,000 annotated Holter recordings. The algorithm generates a summary report including total heartbeats, premature atrial contractions (PACs), premature ventricular contractions (PVCs), couplets, triplets, and runs of tachycardia. For atrial fibrillation, the algorithm calculates the burden (percentage of time in AF). The physician can override automated classifications and add manual annotations. The final report is exported as a PDF and uploaded to the EMR.

Technical Note: The Holter recorder’s 7day battery life is achieved with continuous recording at 250 Hz sampling; increasing the sampling rate to 500 Hz reduces battery life to 48 hours. The solidgel electrodes should be stored in a sealed container; driedout electrodes will cause high impedance and signal noise.

Priorto installs the EEG module as an option on the Patient Monitor platform, or as a standalone EEG system for epilepsy monitoring units. The EEG module provides 32 channels of input, with a sampling rate of 2,000 Hz per channel and a resolution of 0.1 μV. The system includes a photic stimulator (flash frequencies 1–50 Hz) and a video recording system synchronized to the EEG trace. The EEG is integrated with the patient monitor’s ECG channel to detect cardiogenic artifacts.

The EEG module’s electrode cap uses 32 sintered Ag/AgCl electrodes placed according to the 1020 system. Impedance is checked automatically before recording; each electrode must be below 5 kΩ for acceptable signal quality. The system’s spike detection algorithm is configured with a threshold of 2.5 standard deviations above baseline and a minimum duration of 50 ms. For longterm monitoring (24–72 hours), a reduced channel set (19 electrodes) is used to improve patient comfort.

The video recording system uses an infrared camera mounted above the patient’s bed, capturing the entire body. The video and EEG are synchronized using a timecode generator; the system displays both traces on a single screen. When an electrographic seizure is detected, the system automatically saves a 5minute epoch (2 minutes before to 3 minutes after onset) and sends an alert to the epilepsy monitoring dashboard. The EEG report includes a seizure count, a distribution map of spike foci, and a power spectral array.

Technical Note: The 32channel EEG module requires a separate preamplifier headbox connected to the monitor by a shielded 2meter cable. The photic stimulator’s flash intensity should be set to 1 joule per flash; higher intensities may induce seizures in photosensitive patients. Electrode impedance checks should be repeated every 6 hours during longterm monitoring; gel drying causes impedance drift.

Priorto configures the Pulse Oximeter as both a standalone device and a module integrated into the Patient Monitor. The oximeter uses a digital signal processing algorithm that separates arterial pulsations from venous and tissue absorption. The device measures SpO₂ from 0% to 100% with an accuracy of ±2% for 70–100% and ±3% for 50–69%. The pulse rate accuracy is ±1 bpm for rates of 30–250 bpm. The oximeter includes a perfusion index (PI) display, with a PI <0.5% indicating low perfusion and potentially unreliable readings.

For neonatal applications, a soft wraparound sensor is used, with a 10second averaging time to avoid motion artifact. For patients receiving supplemental oxygen, the oximeter’s alarm limits are set by the physician: low SpO₂ alarm typically at 90% for general wards, 88% for COPD patients (targeting SpO₂ 88–92%). The oximeter stores up to 72 hours of trending data, which can be downloaded to a computer via USB. The data includes timestamped SpO₂, pulse rate, and PI every 4 seconds.

The oximeter is integrated with the central monitoring station; when an SpO₂ below the alarm limit persists for 30 seconds, the system sends a notification to the assigned nurse’s mobile device. For sleep studies, the oximeter is configured with a 1second averaging time to capture desaturation events (≥4% drop from baseline). The device’s battery provides 48 hours of continuous operation; a lowbattery alert appears at 20% remaining.

Technical Note: The Pulse Oximeter’s accuracy specification is validated only when the sensor is placed on a finger with adequate perfusion (PI >1%). For patients with hypothermia or vasopressor use, an ear or forehead sensor may provide better signal. Nail polish (especially dark colors) can cause SpO₂ underestimation of up to 5%; removal is recommended.

Priorto provides a central system for monitoring and storing patient data. This system pulls data from several devices. These include patient monitors, Holter monitors, EEG systems, and pulse oximeters.

The system uses a time synchronized database. The accuracy of the network time protocol is plus or minus 10 milliseconds. This aligns all physiological signals.

Hospital staff can access the platform from any workstation. Access is through a secure web interface. The system uses role based access control. Nurses can see current vital signs. Physicians can see full trends and raw waveforms.

The platform has a rules engine. This engine creates alerts for events that involve multiple parameters. For example, bradycardia together with a drop in SpO2 suggests a breathing problem.

For patients wearing a Holter monitor, the platform shows a 24 hour trend. This includes heart rate, ectopy burden, and ST segment deviation.

For EEG patients, the platform provides a seizure detection summary. This summary includes a montage view of all channels during each seizure event.

The platform automatically creates a discharge summary for each patient. This summary includes all monitoring parameters, alarm history, and staff interventions.

All data is stored for 10 years. The storage uses a redundant array of independent disks at level 60.

For research purposes, staff can export de identified data. The export formats are EDF, which stands for European Data Format, or HL7 FHR.

The platform includes a quality dashboard. This dashboard shows monitor utilization, alarm frequency by hospital unit, and staff response times. The clinical engineering committee reviews this dashboard every month.

Technical note: The central platform needs a dedicated server. The server must have at least 32 CPU cores and 128 gigabytes of RAM. Each additional 100 monitored beds adds about 8 gigabytes of RAM.

The NTP time source should be a stratum 1 GPS clock. A stratum 2 clock or lower will cause drift of more than 10 milliseconds after 24 hours.