WO2022118306A1 - Appareil de détection de tumeur de la tête servant à détecter une tumeur de la tête et méthode associée - Google Patents

Appareil de détection de tumeur de la tête servant à détecter une tumeur de la tête et méthode associée Download PDF

Info

Publication number
WO2022118306A1
WO2022118306A1 PCT/IL2021/051320 IL2021051320W WO2022118306A1 WO 2022118306 A1 WO2022118306 A1 WO 2022118306A1 IL 2021051320 W IL2021051320 W IL 2021051320W WO 2022118306 A1 WO2022118306 A1 WO 2022118306A1
Authority
WO
WIPO (PCT)
Prior art keywords
pulse
pulse train
spdw
pulses
ptdw
Prior art date
Application number
PCT/IL2021/051320
Other languages
English (en)
Inventor
Dan SHOMRON
Original Assignee
Shomron Dan
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shomron Dan filed Critical Shomron Dan
Publication of WO2022118306A1 publication Critical patent/WO2022118306A1/fr

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
    • A61B5/316—Modalities, i.e. specific diagnostic methods
    • A61B5/369—Electroencephalography [EEG]
    • A61B5/372—Analysis of electroencephalograms
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/40—Detecting, measuring or recording for evaluating the nervous system
    • A61B5/4058—Detecting, measuring or recording for evaluating the nervous system for evaluating the central nervous system
    • A61B5/4064—Evaluating the brain
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/48—Other medical applications
    • A61B5/4887—Locating particular structures in or on the body
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis

Definitions

  • the invention relates to detection of head tumors and other lesions.
  • Prior art attempts have been made for automating detection of tumors using EEG signals.
  • the prior art attempts can be generally classified into three approaches: band pattern recognition, wavelet transform, and absolute band power.
  • USPA 20190110754 Machine learning based system for identifying and monitoring neurological disorders
  • a signal processing method and system combines multi-scale decomposition, such as wavelet, pre-processing together with a compression technique, such as an auto- associative artificial neural network, operating in the multi-scale decomposition domain for signal denoising and extraction.
  • a compression technique such as an auto- associative artificial neural network
  • USPA 20190307345 Signal processing method for distinguishing and characterizing high-frequency oscillations.
  • a device and a signal processing method that can be used with a device to recognize and distinguish a physiological high- frequency oscillation (HFO) from a pathological high-frequency oscillation.
  • Method applying a wavelet convolution to the electrical signals to generate a first time-frequency representation of the power of the signal.
  • EEG machines include channels for acquiring EEG recordings.
  • Head Tumor Detection Apparatus uses EEG machines include at least 20 channels in accordance with international 10-20 system.
  • De-interleaving algorithms process single pulses to generate pulses trains.
  • a suitable de-interleaving algorithm is described in “De-Interleaving and Identification of Pulsed Radar Signals Using ESM Receiver System”, Kumar et al, International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, Vol. 4, Issue 7, July 2015, pages 6243-6252 incorporated herein by reference.
  • EEG recordings are typically in the order of 20 minutes to 60 minutes. Long EEG recordings can be several days. EEG recordings are typically divided into 3 second epochs. EEG recordings are acquired by commercially available EEG machines which typically have between 20 to 30 channels in accordance with 10/20 international standard. EEG recordings are commonly processed for displaying a time-based voltage for analysis by neurologists for detection of epilepsy, seizures, and head lesions.
  • ERP Effective Radiated Power
  • ERP is the maximum pulse power of a single pulse.
  • EEG epoching is a procedure in which specific time-windows are extracted from a continuous EEG signal. These time windows are called “epochs”. Typically, an EEG epoch is 3 seconds depending on EEG test, recording type and equipment type.
  • Standard EEG Frequency bands are as follows: 1 Hz - 4 Hz Delta band; 4 Hz - 8 Hz Theta band, 8 Hz -13 Hz Alpha band, 13 Hz -30 Hz Beta band, and 30 Hz - 40 Hz Gama band.
  • the present invention employs high-definition frequency bands having a bandwidth of 1Hz to 2Hz.
  • the present invention relates to 37 frequency bands in the 1Hz to 40 Hz band range and additional 37 frequency bands in the 61 Hz to 100 Hz band range, if available.
  • the present invention preferably employs overlapping bands for more accurate test results. Exemplary overlapping bands are 1-3 Hz (Bandl), 2-4Hz (Band2), 3-5Hz (Band3), etc.
  • Epoch groups are groups of time consecutive epochs.
  • An epoch group typically contains between 9 to 50 epochs.
  • Head lesions describe damage or destruction to any part of the head, trauma to the head, and certain health conditions. Malignant or benign tumors are all considered head lesions. Lesion types include inter alia vascular lesions, elevated Intracranial Pressure (ICP), hydrocephalus, Traumatic Brain Injury (TBI), inflammation / infection (meningitis), necrotic tissue, hemorrhagic and even neuro degenerative processes.
  • ICP Intracranial Pressure
  • TBI Traumatic Brain Injury
  • Meningitis inflammation / infection
  • necrotic tissue necrotic tissue
  • hemorrhagic even neuro degenerative processes.
  • the present invention is described for head tumor detection but is equally applicable for head lesion detection.
  • the present invention employs the conventional Talairach or MNI coordinate systems for head tumor location.
  • Head tumors originate from head cells or those that metastasize from other organs that can affect head function in two ways.
  • a head tumor can destroy head cells so that their function is lost or a head tumor can take up space and cause pressure and swelling that affects head cell function. This may occur with benign or cancerous head tumors.
  • Common head tumors include inter alia meningiomas, adenomas, pituitary adenomas, glioma, glioblastoma multiforme, and the like.
  • Normal EEG is an EEG recording of a person that indicates that he does not have a head lesion.
  • the NR database includes normal range tables for various tests divided into single pulse tests and pulse train tests. Each normal range table includes normal EEG limits for various parameters for a particular test. The NR database is used to determine whether an EEG recording is a “Normal EEG” or a “Suspected EEG”.
  • a pulse is classified as a “Normal Pulse” when its Single Pulse Description Word (SPDW) values for a particular single pulse test are within the test’s normal range.
  • SPDW Single Pulse Description Word
  • a pulse train is classified as a “Normal Pulse Train” when its Pulse Train Description Word (PTDW) values for a particular pulse train test are within the test’s normal range.
  • PTDW Pulse Train Description Word
  • PSD Power spectral density
  • Pulse Repetition Interval is the elapsed time from the maximum power of one pulse to the maximum power of its consecutive pulse.
  • PRI is the reciprocal of commonly used pulse repetition frequency (PRF).
  • Pulse trains include all single pulses assumed to be generated by a single source. Pulse trains can include single pulses from one or more frequency bands and/or one or more channels and/or one or more epochs.
  • PTDW data set includes Pulse Train Description Words (PTDWs) of an EEG recording.
  • PTNR data set includes tables with the parameters of pulse train normal range values.
  • PTPRI is the PRI of a pulse train.
  • PTPRI depends on a pulse train PRI type.
  • PTPRI is the time interval between 2 consecutive pulses.
  • PTPRI is the average time between pulses.
  • PTPRI is the time between the first pulses of two consecutive groups of pulses.
  • PTPRI is the time interval between 2 consecutive pulses in the same group.
  • Pulse Width (PW) PW
  • the Pulse Width is the time between the 3 dB points of a single pulse’s maximum power. 3 dB is equivalent to ERP/2.
  • Single pulse extraction involves processing a signal to extract one or more single pulses in an extraction period, if any.
  • Single Channel Pulse Train is a pulse train from a single channel.
  • Single Channel Pulse Trains include at least one single pulse.
  • SCPTDW Data set includes Single Channel Pulse Train Description Words (SCPTDWs) of an EEG recording.
  • Single pulse is a sinusoidal-amplitude pulse and frequency carrier truncated by a symmetrical rectangular-like function of width.
  • single pulses are extracted after high-definition 1Hz to 2Hz band pass filtering of PSD data derived from an EEG recording.
  • SPDW Single Pulse Description Word
  • SPDW Single Pulse Description Word
  • SPDW data set includes Single Pulse Description Words (SPDWs) of an EEG recording.
  • SPNR Pulse Normal Range
  • SPNR data set includes tables with the parameters of Single Pulse normal range values.
  • a pulse is classified as a “Suspected Pulse” when its SPDW values for a particular single pulse test are beyond the test’s normal range.
  • a pulse train is classified as a “Suspected Pulse Train” when its PTDW values for a particular pulse train test are beyond the test’s normal range.
  • the present invention is directed towards a Head Tumor Detection Apparatus (HTDA) for detecting a head tumor using waveform analysis on time-based EEG raw data.
  • the HTDA is a diagnostic tool that includes three components: a commercially available EEG machine, a Head Tumor Detection Processor (HTDP) for processing fresh EEG recordings and a Normal Range database.
  • the HTDA is based on the histological findings that head tumors have different electrical properties and their resistivity values can be used to distinguish tumor tissue from surrounding healthy tissue and to identify tumors and classify certain head tumor types.
  • the HTDA is based on a belief that some head tumors block waveforms and others deflect them or amplify them, and furthermore, some head tumors may be pulse sources for emitting electrical pulses in a similar manner to healthy tissue.
  • the present invention is based on the notion that so-called high-definition single pulses can include a single pulse characteristics indicative of a head tumor and/or a combination of two or more single pulse characteristics indicative of a head tumor.
  • pulse trains of high-definition single pulses can include a single pulse train characteristic indicative of a head tumor and/or a combination of two or more pulse train characteristics indicative of a head tumor.
  • the present invention has been reduced to practice after an extensive iterative process of empirical analysis of over 4,000 EEG recordings of healthy patients and head tumor patients classified into three types: Normal EEG recordings, Head tumor patients and Head lesion patients to determine the following:
  • SPDW Single Pulse Description Word
  • the HTDP initially employs waveform analysis to extract single pulses from an EEG recording per channel per frequency band for a multitude of channels and a multitude of frequency bands. Frequency bands can range from a single Hertz frequency to several Hertz frequency.
  • the HTDP can extract single pulses from a wide range of standard EEG raw data formats including inter alia eeg, ent, edf, EDF+, etc.
  • the HTDP preferably extracts single pulses for typically 25 channels and 74 frequency bands. EEG recordings are typically divided into 3 second epochs. An epoch may, or may not, include single pulses which are used for the analysis.
  • the HTDP prepares a Single Pulse Description Word (SPDW) for each single pulse. Each single pulse is tested in a series of single parameter tests and multiple parameters tests comparing a tested single pulse parameters to normal range values.
  • SPDW Single Pulse Description Word
  • the HTDP also de-interleaves single pulses into pulse trains.
  • the HTDP typically forms thousands of pulse trains from a 5 minute EEG recording.
  • the HTDP prepares a Pulse Train Description Word (PTDW) for each pulse train for a series of pulse train tests comparing a tested pulse train against normal range values.
  • PTDW Pulse Train Description Word
  • the HTDP compares the SPDWs and the PTDWs to the Normal Range database to arrive at a likelihood of a head tumor.
  • a single pulse can lead to a positive test result due to one or more of its parameters. Different single pulse tests are intended to test different single pulse parameters.
  • a pulse train can lead to a positive test result due to one or more of its parameters. Different pulse train tests are intended to test different pulse train parameters. Testing an EEG recording can involve tens of thousands and even hundreds of thousands of test points. Each single pulse test and/or pulse train test has a relatively low probability to predict an occurrence of a head tumor but aggregation of test results affords a high likelihood of detection of a head tumor, if present.
  • the HTDA can also be implemented for providing an approximate location of a head tumor and its type.
  • the HTDA can be employed for detecting head tumors in real time and offline.
  • the present invention can be implemented as a cloud resource for offline processing.
  • the present invention can be implemented in standalone devices in a clinic, outpatient department, and the like.
  • Presently deployed EEG devices for acquiring EEG recordings can be upgraded in accordance with the present invention for detecting head tumors.
  • Fig. 1 shows Head Tumor Detection Apparatus (HTDA) for detecting a head tumor
  • Fig. 2 is a table of clinical test results
  • Fig. 3 is a schematic representation of the international 10-20 system
  • Fig. 4 shows a table of channel descriptions and labels of the international 10-20 system and their 3-D positions in accordance with the MNI and Talairach coordinate systems;
  • Fig. 5A is an exemplary screen shot showing voltage graph of a 3 second epoch EEG recording in a conventional 13Hz - 30Hz Beta frequency band;
  • Fig. 5B is an exemplary screen shot showing a power graph and a frequency graph time synchronized with the Figure 5 A voltage graph;
  • Fig. 5C is a n exemplary screen shot showing PSD data multi-graph of the same 3 second epoch EEG recording in a high-definition 24Hz-26Hz frequency band;
  • Fig. 6 is a top-level flow diagram of the operation of the head tumor detection apparatus for outputting a patient report
  • Fig. 7 is a part of a Single Pulse NR Table FreqPW
  • Fig. 8 is a part of a Single Pulse NR Table FreqDirM
  • Fig. 9 is a part of a Pulse Train NR Table FreqPRI;
  • Fig. 10 is a part of a Pulse Train NR Table Stagger
  • Fig. 11 is a part of the SPDW data set showing three exemplary SPDWs
  • Fig. 12 is a top level flow diagram of waveform processing of EEG raw data to build the Single Pulse Description Word data set;
  • Fig. 13 is a flow diagram of the extraction of single pulses from an EEG recording
  • Fig. 14 is a n exemplary screen shot showing PSD data multi-graph of an epoch including three single pulses;
  • Fig. 15A is a n exemplary screen shot showing power graph showing the finding of the strongest power in the power graph of the Figure 14 epoch;
  • Fig. 15B is an exemplary screen shot showing power graph showing pulse extraction of a first single pulse of the Figure 14 epoch
  • Fig. 15C is an exemplary screen shot showing power graph showing pulse extraction of a second single pulse of the Figure 14 epoch
  • Fig. 15D is an exemplary screen shot showing power graph showing power graph showing pulse extraction of a third single pulse of the Figure 14 epoch;
  • Fig. 15E is an exemplary screen shot showing power graph showing power graph showing attempt of pulse extractions from the Figure 14 epoch;
  • Fig. 16A is the same as Figure 14 showing three extracted single pulses
  • Fig. 16B shows a part of the SPDW data set including the SPDWs for the extracted three Figure 16A single pulses
  • Fig. 17 is exemplary screen shot showing PSD data multi- graph of an epoch including five pulses
  • Fig. 18 shows a part of SPDW data set including the SPDWs for the five Figure 17 single pulses
  • Fig. 19 is a top level flow diagram for processing a SPDW data set to build a PTDW data set
  • Fig. 20A shows a part of the SCPTDW data set showing five exemplary SCPTDWs
  • Fig. 20B shows a part of the PTDW data set showing four exemplary PTDWs
  • Fig. 21 is an exemplary screen shot showing a PSD data multi-graph of a pulse train classified as a constant PRI type
  • Fig. 22 shows a schematic representation of the Figure 21 pulse train’s PTDW
  • Fig. 23 is an exemplary screen shot showing a PSD data multi-graph of a pulse train classified as a stagger PRI type
  • Fig. 24 shows a schematic representation of the Figure 23 pulse train’s PTDW
  • Fig. 25 is an exemplary screen shot showing a PSD data multi-graph of a pulse train classified as a jitter PRI type
  • Fig. 26 is a schematic representation of the Figure 25 pulse train’s PTDW;
  • Fig. 27 is an exemplary screen shot showing a PSD data multi-graph of a pulse train classified as a pulse group PRI type;
  • Fig. 28 is a schematic representation of the Figure 27 pulse train’s PTDW;
  • Fig. 29A is a top-level flow diagram for testing the SPDW data set
  • Fig. 29B shows a part of the SPDW data set showing three exemplary SPDWs
  • Fig. 29C shows use of the Single Pulse Normal Range Table for testing the Figure 29B SPDWs
  • Fig. 30A is a flow diagram of a Single Pulse Test FreqPW
  • Fig. 30B shows a part of the SPDW data set showing three exemplary SPDWs
  • Fig. 30C shows use of the Single Pulse Normal Range Table FreqPW for testing the Figure 30B SPDWs;
  • Fig. 31 is a flow diagram of a Single Pulse Test FreqDirM
  • Fig. 3 IB shows a part of the SPDW data set showing an exemplary SPDW
  • Fig. 31C shows use of the Single Pulse Normal Range Table FreqDirM for testing the Figure 3 IB SPDW;
  • Fig. 32 is a top-level flow diagram for testing the PTDW data set
  • Fig. 33A is a flow diagram of a Pulse Train Test FreqPRI;
  • Fig. 33B shows a part of the PTDW data set showing four exemplary PTDWs
  • Fig. 33C shows use of the Pulse Train Normal Range Table FreqPRI for testing the Figure 33B PTDWs
  • Fig. 34A is a flow diagram of a Pulse Train Test Stagger
  • Fig. 34B shows a part of the PTDW data set showing three exemplary PTDWs
  • Fig. 34C shows use of the Pulse Train Normal Range Table Stagger for testing the Figure 34B PTDWs
  • Fig. 35A is a flow diagram of location calculation
  • Fig. 35B shows an exemplary PTDW
  • Fig. 35C shows four exemplary SCPTDWs of the Figure 35B PTDW
  • Fig. 35D shows a graphic representation of the location of a head tumor.
  • the present invention may be a system, a method, and/or a computer program product.
  • the computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.
  • the computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device.
  • the computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
  • a non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD- ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, and any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read-only memory
  • EPROM or Flash memory erasable programmable read-only memory
  • SRAM static random access memory
  • CD- ROM compact disc read-only memory
  • DVD digital versatile disk
  • memory stick a floppy disk, and any suitable combination of the foregoing.
  • a computer readable storage medium is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
  • Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network.
  • the network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers.
  • a network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
  • Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the “C” programming language or similar programming languages.
  • the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.
  • the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
  • electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
  • These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
  • the computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
  • each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s).
  • the functions noted in the block may occur out of the order noted in the figures.
  • two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.
  • Section 4 Testing for Head Tumor Detection and Location
  • FIG. 1 shows Head Tumor Detection Apparatus (HTDA) 10 for detecting a head tumor.
  • the HTDA 10 includes the following three components:
  • a standard EEG machine 11 for obtaining EEG raw data from patients is provided.
  • a Head Tumor Detection Processor (HTDP) 12 for processing EEG raw data for extracting single pulses and forming pulse trains.
  • the HTDP 12 determines Single Pulse Description Words (SPDWs) for single pulses and Pulse Train Description Words (PTDWs) for pulse trains.
  • SPDWs Single Pulse Description Words
  • PTDWs Pulse Train Description Words
  • the HTDP 12 runs a series of single pulse tests and pulse train tests to arrive at a likelihood of a head tumor.
  • single pulse extraction and pulse train formation are described for 3 second epochs which is a common standard EEG recording format.
  • the teachings of the present invention can be equally applied to shorter or longer epochs.
  • the teachings of the present invention can also be equally applied to a single epoch for an entire EEG recording.
  • a Normal Range database 13 includes Single Pulse Normal Range Tables 15 for a series of single pulse tests and Pulse Train Normal Range Tables 16 for a series of pulse train tests.
  • the single pulse tests are intended to test different single pulse parameters including inter alia ERP, Frequency, PW, Rise Time, phase changes within pulses, pulse symmetry, etc.
  • Pulse train tests are intended to test different pulse train parameters including inter alia PRI type, PRI values, pulses that are received in multiple frequencies, pulses that are received in multiple channels, etc.
  • the HTDA 10 includes an interface to receive EEG recordings from the EEG machine 11 and can be in communication with a hospital medical system.
  • the HTDA 10 includes a user interface for inputting patient information and a display screen for displaying EEG recordings, test results, patient reports 14, and the like.
  • Figure 2 is a table of clinical test results achieved by the HTDA 10 for 2030 EEG recordings.
  • the HTDA 10 achieved the following impressive clinical results: All Head Tumor patients were detected. Normal EEG recordings had minor false positive detections of less than 2%. Head lesion patients had minor false negative detections of 1.2%.
  • FIG. 3 shows the EEG international 10-20 system.
  • Figure 4 shows a table of channel descriptions and labels according to the international 10-20 system and their 3-D positions in accordance with the MNI and Talairach coordinate systems.
  • the HTDA 10 employs these channel labels for indicating an approximate location of a head tumor in a patient report.
  • Figure 5A is a voltage graph of a 3 second epoch EEG recording in the standard 13Hz - 30Hz Beta band as obtained in conventional EEG analysis approaches.
  • Conventional EEG analysis approaches uses a spectral method to aggregate the epoch power and identify the main epoch frequency in the standard 13Hz - 30Hz Beta band.
  • Figure 5B shows a power graph and a frequency graph time synchronized with the Figure 5 A voltage graph.
  • Figure 5B shows a major 15Hz emission with Time of Arrival (TO A) 1.3 seconds and an additional minor 15Hz emission at Time of Arrival (TOA) 2.2 seconds.
  • TO A Time of Arrival
  • TOA Time of Arrival
  • the HTDP 12 processes EEG recordings for extracting single pulses in high-definition frequency bands obtained using narrow band pass filters of between 1 Hz to 2Hz.
  • Figure 5C shows PSD data multi-graph obtained subsequent to waveform processing according to the present invention for the 24 Hz - 26 Hz frequency band.
  • the PSD data multi-graph shows two about 25 Hz single pulses with TOAs 1.3 sec and 2.2 sec, respectively. These single pulses are concealed by stronger emissions in different frequencies when using conventional EEG frequency bands but are important for head tumor detection as described hereinbelow.
  • FIG. 6 shows HTDA operation includes the following steps:
  • the HTDP 12 performs the following system initializations:
  • the HTDP 12 uses a default head coordinate system: the Talairach head coordinate system.
  • the default channel locations are set according to the standard 10-20 international channel location system as based on “Three- dimensional probabilistic anatomical cranio-cerebral correlation via the international 10-20 system oriented for transcranial functional brain mapping.” Masako Okamoto, a,l Haruka Dan, a,l Kuniko Sakamoto, a Kazuhiro Takeo, b Koji Shimizu, b Satoru Kohno,b Ichiro Oda, c Seiichiro Isobe, a Tateo Suzuki, a Kaoru Kohyama,a and Ippeita Dana, NeuroImage Volume 21, Issue 1, January 2004, Pages 99-111”.
  • the HTDP 12 sets default Minimum Power Thresholds (MPTs) per frequency. For example, for the 3Hz frequency band, a typical default MPT value is 5 e+03 ⁇ W.
  • the HTDP 12 sets the default HMI format of the default patient report 14
  • Step 52 Obtain EEG recording.
  • An EEG recording is either received online or read from a file.
  • the HTDA 10 supports multiple types of conventional EEG formats including inter alia eeg, ent, edf, EDF+, etc.
  • Step 54 The HTDP 12 performs waveform processing for pulse extraction from the EEG recording.
  • the HTDP 12 determines a Single Pulse Description Word (SPDW) per each extracted pulse per channel per frequency band.
  • the HTDP 12 stores the SPDWs in a SPDW data set 56.
  • the waveform processing is described hereinbelow in greater detail in Section 2: Single Pulse Extraction and Determination of Single Pulse Description Words (SPDWs).
  • Step 58 De-interleaving the SPDW data set 56 to generate pulse trains.
  • the HTDP 12 determines a Pulse Train Description Word (PTDW) per pulse train.
  • the HTDP 12 stores the PTDWs in a PTDW data set 60.
  • the de- interleaving is described hereinbelow in greater detail in Section 3: Pulse Train Formation and Determination of Pulse Train Description Words (PTDWs).
  • Step 62 Head tumor detection and location includes testing the SPDW data set 56 using the Single Pulse Normal Range tables to achieve test results for each single pulse and the PTDW data set 60 using the Pulse Train Normal Range tables to achieve test results for each pulse train. Single pulse tests and pulse train tests are designed such that detection of a head tumor leads to a positive test result.
  • Head tumor detection includes aggregating test results to prepare a patient report including a likelihood of a head tumor.
  • the patient report also preferably includes an approximate 3D location of a head tumor in Talairach coordinates.
  • FIG. 7 shows part of a Single Pulse Normal Range Table FreqPW for testing single pulses as described hereinbelow in Section 4: Testing for Head Tumor Detection and Location.
  • the Single Pulse Normal Range table FreqPW includes the following columns:
  • FIG. 8 shows part of a Single Pulse Normal Range Table FreqDirM for testing single pulses as described hereinbelow in Section 4: Testing for Head Tumor Detection and Eocation.
  • the Single Pulse Normal Range Table FreqDirM includes the following columns:
  • Figure 9 shows part of a Pulse Train Normal Range Table FreqPRI for testing pulse trains as described hereinbelow in Section 4: Testing for Head Tumor Detection and Location.
  • the Pulse Train Normal Range Table FreqPRI includes the following columns:
  • Figure 10 shows a Pulse Train Normal Range Table Stagger for testing pulse trains as described hereinbelow in Section 4: Testing for Head Tumor Detection and Location.
  • the Pulse Train Normal Range Table Stagger includes the following columns:
  • FIG 11 shows three exemplary SPDWs of the SPDW data set 56.
  • Each SPDW includes the following significant fields:
  • Field 1 Pulse Id used for database management.
  • Test Id representing EEG Recording Id. This field is employed when multiple EEG recordings are combined
  • Field 6 Pulse in epoch: The one or more pulses extracted from an epoch are sorted by their Time of Arrival (TOA)
  • TOA Time of Arrival
  • Field 8 ERP: The maximum pulse power of a single pulse
  • Pulse Width is the time between the 3 dB points of a single pulse’s maximum power. 3 dB is equivalent to ERP/2.
  • Field 12 Symmetric: Logical flag having either TRUE or FALSE value according to the following relationship: where the symmetric factor is typically set to 50% of ERP. The symmetric factor calculation is affected by the relative ERP and scan rate that are hardware dependent and can be set to values between 25% to 75% of the ERP. SPDWs can include additional single pulse parameters not shown in Figure 11. Exemplary additional single pulse parameters include inter alia Max Phase Jump: Maximum change of phase within a pulse
  • Min Phase Jump Minimum change of phase within a pulse
  • Rise Time Time from minimal pulse power to reach a predetermined percentage of maximum pulse power.
  • the predetermined percentage is typically 80%.
  • Figure 12 shows a flow diagram of waveform processing of EEG raw data to determine a Single Pulse Description Word (SPDW) per pulse per channel per frequency band.
  • the waveform processing includes the following steps:
  • Step 80 Start waveform processing for pulse extraction
  • Step 82 Employ Fast Fourier Transform (FFT) to convert EEG raw data to frequency domain
  • Step 84 Initial frequency filtering typically in a range of from 0.5 Hz to about 40 Hz range.
  • Initial frequency filtering can also include an additional range from 60.5 Hertz and above. The additional range depends on EEG machine type, electricity infrastructure in a country where an EEG recording is being obtained, and the like.
  • Step 86 Using a spectral method to calculate Power Spectral Density (PSD) for estimating the power of the EEG recording at different frequencies to prepare PSD data.
  • Suitable spectral methods include inter alia Welch’s method, and the like.
  • Step 88 High- Definition (HD) frequency filtering of the PSD data for subsequent pulse extraction.
  • the HD frequency filtering uses narrow band pass filters of typically 2 Hz bandwidth with 1 Hz overlapping between adjacent frequency bands. For example, Bandl (1-3Hz), Band 2 (2-4Hz), Band3 (3-5Hz), etc.
  • the HD filtering employs 37 band pass filters for the 0.5 Hz to about 40 Hz range and additional 37 band pass filters for the 60 Hz to 100 Hz range when available.
  • Step 90 Extract pulses from PSD data per frequency band per channel.
  • Single pulse extraction can be performed using different approaches.
  • Figure 13 to Figure 16 show one approach of single pulse extraction from a 3 second epoch EEG recording. Pulse extraction is preferably performed per epoch as opposed to an entire EEG recording or part thereof.
  • Step 92 Determining a SPDW for each extracted pulse for storing in the SPDW data set 56
  • Figure 13 shows single pulse extraction from an EEG recording. The single pulse extraction is performed per epoch per channel per frequency band.
  • Step 100 Start
  • Step 102 Obtain filtered PSD data of one epoch and one channel
  • Step 104 Find strongest power in the filtered PSD data between start of epoch and end of epoch
  • Step 106 Calculate Minimum Power Threshold (MPT) which is the maximum of either Step 50’s pre-determined MPT or Step 104’s strongest power found divided by ThreshDividor where a typical value of ThreshDividor is 2.5
  • Step 108 Set extraction period start time to 0 second
  • Step 110 If power at the start time exceeds MPT, then this point is considered the start time of a single pulse, otherwise search for the first point where the power upwardly crosses the MPT. If found, continue to Step 112 and otherwise continue to Step 126 End.
  • Step 112 Set the point to pulse start time
  • Step 116 Set point found to be the pulse end time
  • Step 118 Write all pulse parameters to SPDW data set 56 as described hereinbelow in greater detail with reference to Figure 14 to Figure 18
  • Step 120 If the pulse end time is 3 seconds, end of epoch, continue to Step 126 End. Otherwise continue to Step 124: Set extraction period start time to pulse end time
  • Step 122 Set pulse end time to 3 sec and continue to Step 118 Write all pulse parameters to SPDW data set 56
  • Step 124 Set extraction period start time to pulse end time and continue to Step 110
  • FIG. 14 shows a PSD data multi-graph with 5 time synchronized graphs: voltage, power, frequency, residual phase and phase.
  • the PSD data multi-graph header includes the following significant fields: an EEG identifier, a channel index, an epoch number and a band pass filter. In the shown example, the significant fields have the following values:
  • EEG identifier EDF-00007750_s001_t000.edf (recording file name and format)
  • Band pass filter [3 - 4.33] Hz.
  • a start frequency of a band pass filter is assigned to a SPDW’s Band field.
  • FIG. 15A to Figure 15E show pulse extraction of three single pulses from the Figure 14’s power graph as follows.
  • Figure 15A shows finding the strongest power in the power graph.
  • the calculated MPT is greater than the default 3 Hz frequency band’s MPT 5 e+06 ⁇ W and therefore the HTDP 12 uses the calculated MPT going forward.
  • the power downwardly crosses the MPT at X 1.444 sec. This is the first pulse’s end time.
  • the SPDW is written to the SPDW data set 56.
  • the power downwardly crosses the MPT at X 1.976 sec. This is the second pulse’s end time.
  • the SPDW is written to the SPDW data set 56.
  • the power downwardly crosses the MPT at X 2.612 sec.
  • the SPDW is written to the SPDW data set 56.
  • Figure 16A shows the Figure 14 PSD data multi-graph with the extracted three pulses shown on the power graph.
  • Figure 16B shows the SPDWs for the three extracted single pulses sorted by their Time of Arrival (TO A).
  • TO A and ERP values are directly taken from the pulse extraction.
  • the MinFreq, MaxFreq, PW and Symmetrical Flag values are determined as described hereinabove.
  • Figure 17 shows a PSD data multi-graph with five single pulses.
  • Figure 18 shows the SPDWs for the five Figure 17 single pulses.
  • the first and last single pulses have a Symmetric flag FALSE because they are at the epoch’s edges.
  • Pulse train formation regards epochs within their time-domain vicinities which can range from several backward consecutive epochs and/or forward consecutive epochs to an entire EEG recording. Pulse train formation is intended to detect pulse trains which can extend along many epochs and can contain important information possibly identifying a tumor. Pulse trains can include pulses which were not tested as single pulses, for example, for failing to be symmetrical. Pulse trains are preferably formed from single channel pulse trains for sake of computational simplicity. Pulse trains are formed from clustering single channel pulse trains that have a common parameter as described hereinbelow.
  • Single channel pulse trains and pulse trains have similar description words except for a few exceptions as follows.
  • Single channel pulse train description words include a list of pointers to single pulses in the SPDW data set 56.
  • Pulse train description words include a list of pointers to single channel pulse trains in a Single Channel Pulse Train Description Word (SCPTDW) data set 160.
  • SCPTDW Single Channel Pulse Train Description Word
  • Figure 19 shows the process for building a PTDW data set 60 from a SPDW data set 56.
  • the process includes the following steps:
  • Step 152 Select epochs for de-interleaving from the SPDW data set 56.
  • Step 154 Update the SPDW data set 56 for epoch edges. If a pulse end time coincides with an epoch end time of an epoch, and in a next epoch, a first pulse start time is 0, then the two non- symmetrical pulses are united as a new single pulse of the earlier epoch and becomes its last single pulse.
  • Step 156 Select the first channel in all frequency bands in the selected epochs for de-interleaving.
  • Step 158 De-interleave the SPDWs to output a Single Channel Pulse Train Description Word (SCPTDW) data set 160
  • Step 162 If this is the last channel, continue to Step 164, otherwise continue to Step 166 Select next channel
  • Step 164 Cluster Single Channel Pulse Trains (SCPTs) to pulse trains which can include SCPTs from multiple channels and output a Pulse Train Description Word (PTDW) for each pulse train to the PTDW data set 60.
  • SCPTs Cluster Single Channel Pulse Trains
  • PTDW Pulse Train Description Word
  • Step 166 Select the next channel and continue to Step 158 De-interleave.
  • Figure 20A shows four exemplary SCPTDWs of a SCPTDW data set 160. The parameters are divided into 2 groups: identifier fields and calculated Single Channel Pulse Train data fields.
  • Test Id representing EEG Recording Id . This field is employed when multiple EEG recordings are combined
  • Field 2 Pulse Train ID, in case of SCPTDW its Single Channel PTDW Id.
  • Field 3 Main Channel - The channel with the strongest pulse in a pulse train
  • Field 4 Number of channels: In the SCPTDW data set, this number is always 1. After the clustering stage, a pulse train will typically have several channels. In the PTDW the field includes the number of channels that have pulses in a pulse train
  • Field 5 Number of Pulses -Number of pulses in this pulse train, in a single channel and a single frequency band.
  • Field 6 Main Band - The band with the strongest pulse in a pulse train
  • Field 7 Number of Bands- A pulse train can have pulses from several frequency bands. This field is the number of frequency bands that have pulses in a pulse train.
  • PTPRI is PRI for a pulse train. PTPRI depends on a pulse train
  • PTPRI is the time interval between 2 pulses.
  • PTPRI is the average time between pulses.
  • PTPRI is the time between the first pulses of two consecutive groups.
  • PTPRI is the time interval between 2 consecutive pulses in the same group.
  • Field 9 PRI Type classified into one of the following four types: constant, pulse group, stagger, and jitter, PRI types are detailed herein in Figures 21-28.
  • Stagger level Stagger is defined as a repetitive pattern of pulses and stagger level is the number of pulses in a pattern.
  • Figure 22 shows two repetitive patterns each including 6 pulses so the Stagger Level field is assigned 6. Such repetitive patterns of pulses are rare and highly indicative of a head tumor.
  • Min PRI is applicable for jitter PRI type pulse trains only. MinPRI is the minimal time between two consecutive pulses in a pulse train.
  • Max PRI is applicable for jitter PRI type pulse trains only. MaxPRI is the maximal time between two consecutive pulses in a pulse train.
  • Field 13 GRI is applicable for pulse group PRI type pulse train only. GRI is the time between the TOA of the first pulse in the first pulse group and the first pulse in the second pulse group.
  • Frequency Hopper flag A flag indicating if a pulse train includes pulses of different frequencies either within a single frequency band or in multiple frequency bands.
  • Fields 21 onwards List of pointers to single pulse description words in the SPDW data set 56.
  • a SCPTDW length is set by its number of pointers to single pulses.
  • FIG. 20B shows four exemplary PTDWs of the PTDW data set 60.
  • Pulse trains are formed from clustering single channel pulse trains that have a common parameter. Suitable common parameters for clustering Single Channel Pulse Trains include inter alia PRI, PRI type, and the like.
  • Each PTDW includes a list of pointers to the SCPTDWs that formed the pulse train enabling to use both the SCPTDW parameters and the single pulses included in a pointed SCPTDW.
  • the first column in the PTDW data set 60 describes a pulse train clustered from SCPTDW data set 160’s columns 1, 3 and 4 which have the following common parameters: PTPRI value 750, PRI type Jitter and Nm of Pulses 33.
  • the second column in the PTDW data set 60 describes a pulse train clustered from SCPTDW data set 160’s columns 2 and 5 which have the following common parameters: PTPRI value 420, PRI type Constant and Nm of Pulses 4.
  • Figure 21 shows a constant PRI type pulse train defined as having single pulses of substantially similar time intervals between adjacent pairs of consecutive pulses.
  • the pulse train has 12 single pulses.
  • Figure 22 shows the PTPRI value is 425 milliseconds.
  • the MinPW and MaxPW are both 160 milliseconds.
  • a pulse train with 3 or more consecutive pulses with substantially similar time intervals between a pair of consecutive pulses is considered a constant PRI type pulse train.
  • Figure 23 shows a stagger PRI type pulse train defined as a repetitive pattern of single pulses in time domain.
  • the pulse train has a repetitive pattern of 6 pulses as marked in the figure.
  • the PTPRI is defined as the time interval between the first and last pulse in a repetitive pattern.
  • Figure 24 shows in the present example, PTPRI is 3410 milliseconds and the Stagger level is 6.
  • Figure 25 shows a jitter PRI type pulse train defined as a pulse train with changing PRI.
  • a pulse train with at list 6 single pulses with approximately the same PW is considered a jitter PRI type pulse train.
  • Figure 26 shows the pulse train has 36 single pulses.
  • Figure 27 shows a pulse group PRI type pulse train defined as a pulse train having at least two identical groups each having at least two single pulses with equal time intervals between them.
  • Figure 27 shows two identical groups of 4 pulses each in epoch 22 and epoch 36.
  • Figure 28 shows the PTPRI value is 730 milliseconds between consecutive pulses within a pulse group.
  • the GRI value between the two pulse groups is 70500 milliseconds.
  • Such pulse groups are rare and highly indicative of a head tumor. Section 4: Testing for Head Tumor Detection and Location
  • Testing for head tumor detection involves testing a patient’s SPDWs and PTDWs.
  • SPDWs are tested on one or more single pulse tests.
  • a single pulse test can be stopped in the event that a single pulse results in a positive test result.
  • all single pulses can be tested after a single pulse has resulted in a positive test result for additional information.
  • the results of two or more single pulse tests can be weighted differently depending on a test’s importance.
  • PTDWs are tested on one more pulse train tests.
  • a pulse train test can be stopped in the event that a pulse train results in a positive test result.
  • all pulse trains can be tested after a pulse train has resulted in a positive test result for additional information.
  • the results of two or more pulse train tests can be weighted differently depending on a test’s importance.
  • Patient reports are generated by aggregation of single pulse test results and pulse train test results.
  • Figure 29A shows a flow diagram for testing the SPDW data set 56. Each single pulse test is preferably performed on all SPDWs.
  • Step 180 Start testing
  • Step 182 Obtain a first SPDW from the SPDW data set 56
  • Step 184 Perform a single pulse test on the SPD.
  • a single pulse test can test either a single parameter or multiple parameters
  • Step 186 If last SPDW in SPDW data set, then continue to Step 188 Last single pulse test? otherwise continue to Step 192 Obtain next SPDW.
  • Step 188 If last single pulse test continue to Step 190: Head tumor detection, otherwise continue to Step 194 Select next single pulse test
  • Step 190 Head tumor detection, aggregate all tests results for head tumor detection
  • Step 192 Obtain next SPDW and return to Step 184
  • Step 194 Select next single pulse test and return to Step 184 Single Parameter Single Pulse Tests
  • a tested SPDW parameter is a less or more than its normal range limit value as shown in Figure 29 C.
  • Each single parameter of each SPDW is tested individually.
  • the tests are carried out in the order of appearance as in Figure 29C.
  • the order of the tests is arbitrary.
  • the first pulse is a Normal Pulse and not Suspected Pulse because all its SPDW’s parameters do not exceed their limits.
  • the second single pulse is also a Normal Pulse because all its SPDW’s parameters do not exceed their limits.
  • the third pulse is a Suspected Pulse because its band 35 ’s rise time 7 msec is less than the band 35’s normal range rise time 8 msec.
  • Condition 1 A tested SPDW symmetric flag is TRUE
  • a tested SPDW ERP is greater than FreqPW’ s power obtained from the band column that its number equals tested SPDW band
  • FIG. 30A shows the Single Pulse Test FreqPW includes the following steps: Step 200: Start FreqPW test
  • Step 202 Obtain first SPDW from SPDW data set 56
  • Step 204 If single pulse is symmetric, continue to Step 206 Find PW row in SPNR Table FreqPW, else continue to Step 216 Last SPDW?
  • Step 206 Find PW row in the SPNR Table FreqPW by checking if there is a row where the single pulse PW is between the SPNR Table FreqPW Minimum PW and Maximum PW
  • Step 208 If PW row found, continue to Step 210 Find band column and read power, else continue to Step 216 Last SPDW?
  • Step 210 Obtain the maximum limit power in the SPNR Table FreqPW PW row and band column that matches the SPDW band
  • Step 212 If SPDW ERP exceeds the Step 210 power, continue to Step 214 Positive Test Result, else continue to Step 216 Last SPDW?
  • Step 214 Add Positive Test Result
  • Step 216 If last SPDW, continue to Step 220 End else continue to Step 218
  • Step 218 Obtain next SPDW and continue to Step 204
  • the SPNR Table FreqPW row 20 has Min PW 210 and Max PW 230 such that the first SPDW falls within row 10 range.
  • the first SPDW ERP value is less than the SPNR Table FreqPW value, so the first single pulse is a Normal Pulse.
  • the second single pulse and the third single pulse are also normal pulses.
  • the Single Pulse Test FreqDirM results in a positive test result in the event that a single pulse’s parameters meet the following conditions:
  • Condition 1 A tested SPDW symmetric flag is TRUE
  • Condition 2 A tested SPDW PW is less than 1000
  • Condition 3 A tested SPDW Channel is between 1 and 22 and its row is in the FreqDirM table
  • a tested SPDW ERP is greater than FreqDirM’ s power obtained from FreqDirM row in which the channel number equals SPDW channel and column equals SPDW Band.
  • Figure 31 A shows the Single Pulse Test FreqDirM includes the following steps:
  • Step 240 Start FreqDirM test
  • Step 242 Obtain first SPDW from SPDW data set 56
  • Step 244 If single pulse is symmetric and SPDW PW ⁇ 1000 milliseconds, then continue to Step 246 Find Chan row, else continue to Step 256 East SPDW?
  • Step 246 Find SPNR Table FreqDirM Chan row that matches SPDW Chan
  • Step 248 If SPNR Table FreqDirM Chan row is found, then continue to Step 250 SPDW. If ERP exceeds FreqDIRM power, else continue to Step 256 East SPDW?
  • Step 250 Find Band column and read power and continue to Step 252
  • Step 252 If SPDW ERP exceeds the SPDW Table FreqDirM power, then continue to Step 254 Add Positive Test Result, else continue to Step 256 East SPDW?
  • Step 254 Add Positive Test Result
  • Step 256 If Last SPDW continue to Step 260, else continue to Step 258
  • Step 258 Obtain next SPDW and continue to Step 244
  • Step 260 End
  • the SPNR Table FreqDirM has encircled value 17703426.
  • the SPDW ERP value is less than the SPNR Table FreqDirM value, so the first single pulse is a Normal Pulse.
  • the SPNR Table FreqDirM has encircled value 9795199.
  • the SPDW ERP value is more than the SPNR Table FreqDirM value, so the second single pulse is not within normal range and results in a positive test result.
  • the positive test result was determined for channel 9, namely, 01 - Left back of the head. The information is considered in a location calculation of a head tumor as described hereinbelow.
  • the SPNR Table FreqDirM has encircled value 886372.3.
  • the SPDW ERP value is less than the SPNR Table FreqDirM value, so the single pulse is a Normal Pulse.
  • Figure 32 shows a flow diagram for testing the PTDW data set 60. Each pulse train test is preferably performed on all PTDWs.
  • Step 270 Start pulse train tests
  • Step 272 Obtain a first PTDW from the PTDW data set 60
  • Step 274 Perform a pulse train test on the PTDW
  • Step 276 If this is not the last PTDW, then continue to Step 278. If last PTDW in PTDW data set, continue to Step 280
  • Step 278 Obtain the next PTDW and continue to Step 274
  • Step 280 Check if the pulse train test is the last pulse train test. If it is the last pulse train test, then continue to Step 282. If it is not the last pulse train test, then continue to Step 284 Select next pulse train test
  • Step 282 Head tumor detection and Location, continue to Step 286
  • Step 284 Select next pulse train test and continue to Step 274
  • Pulse Train Test FreqPRI Pulse Train Test FreqPRI with reference to Figure 33A to Figure 33C and Pulse Train Test Stagger with reference to Figure 34A to Figure 34C. Pulse Train Test FreqPRI
  • Condition 3 A tested pulse train’s PTPRI is between FreqPRI’s Min PRI and Max PRI
  • Figure 33A shows a flow diagram of a Pulse Train Test FreqPRI using the Figure 33C Pulse Train Normal Range (PTNR) Table FreqPRI.
  • PTNR Normal Range
  • Test FreqPRI includes the following steps:
  • Step 290 Start FreqPRI test
  • Step 292 Obtain a first PTDW from the PTDW data set 60
  • Step 294 Test the PTDW in FreqPRI test which includes 4 sub-tests.
  • the FreqPRI test is tested against each of the rows of the Pulse Train NR Table: FreqPRI.
  • the test result is positive if a tested PTDW has a positive result in all sub tests in any of the FreqPRI rows. The following are the sub-tests performed:
  • Sub-test4 PowerTest if PTDW Min ERP exceeds PTNR Table FreqPRI Min ERP, If the results in all sub-tests are positive, than the Pulse Train Test FreqPRI results in a positive test result, continue to Step 296, else continue to Step 300, Last
  • Step 296 Add Positive Test Result and continue to Step 298
  • Step 298 Calculate location using Angle- of- arrival localization detailed herein in Figure 35, and continue to Step 300 Last PTDW?
  • Step 300 If this is the last PTDW continue to Step 304 End else continue to Step 302 Obtain the next PTDW
  • Step 302 Obtain the next PTDW and continue to Step 294
  • Step 304 End
  • Nm of Pulses is 33 which is greater than FreqPRI first row Min Nm of Pulses - positive, PTDW Main Band is 7 same as FreqPRI - positive, PTDW PTPRI is 750 which exceeds Max PRI in first row of FreqPRI table which is 100 so the PTDW is checked versus the next FreqPRI row. Since in all other FreqPRI rows Freq Band is not 7, then the first pulse train is a normal pulse train.
  • Nm of Pulses is 33 which is greater than FreqPRI first row Min Nm of Pulses - positive, PTDW Main Band is 7 and FreqPRI Freq Band is 8 so the test continues to the second row.
  • Nm of Pulses is 4 which is equal FreqPRI first row Min Nm of Pulses - positive, PTDW Main Band is 8 same as FreqPRI Freq Band - positive, PTDW PTPRI is 420 which exceeds Max PRI in second row of FreqPRI table which is 100 so the PTDW is checked versus the next FreqPRI row.
  • the Freq Band is 9 compared to the PTDW main band which is 7 so the tests continue with row 5.
  • PTDW Nm of Pulses is 33 which is greater than FreqPRI fifth row Min Nm of Pulses - positive
  • PTDW Main Band is 8 same as FreqPRI row 5 - positive
  • PTDW PTPRI is 420 which exceeds Max PRI in fifth row of FreqPRI table which is 400 so the PTDW is checked versus the next FreqPRI row. In this example this is the last row tested, so the second pulse train is a normal pulse train.
  • the third pulse is also a normal pulse train since its PTPRI 1200 exceeds the Max PRI in each of the FreqPRI table Max PRI columns.
  • Nm of Pulses is 8 which is greater than FreqPRI first row Min Nm of Pulses - positive
  • PTDW Main Band is 7 same as FreqPRI - positive
  • PTDW PTPRI is 90 which is between FreqPRI first row MIN PRI which is 60 and Max PRI which is 100 - positive
  • PTDW’s Max ERP is 32000 which exceeds FreqPRI first row Min ERP which is 30000. Accordingly, the fourth pulse train is not a normal pulse train since all 4 sub-tests have positive results resulting in a positive test result.
  • the Pulse Train Test Stagger results in a positive test result in the event that a pulse train’s parameters meets one of the following three conditions:
  • Figure 34A shows the Pulse Train Test Stagger includes the following steps: Step 320: Start of stagger test
  • Step 322 Obtain the first PTDW from the PTDW data set 60
  • Step 324 If PTDW PRI Type is Stagger, then continue to Step 326 Stagger level > 3, else continue to Step 338 Last PTDW?
  • Step 326 If PTDW Stagger Level > 3, then continue to Step 328 Stagger level > 4, else continue to Step 330 PRITest Step 328: If PTDW Stagger Level > 4, then continue to Step 334 Add Positive Test Result, else continue to Step 332 FreqTest
  • Step 330 Perform PRITest: if PTDW PRI Is between Pulse Train Normal Range Table Stagger Min PRI and Max PRI, continue to Step 332 FreqTest else continue to Step 338 Last PTDW?
  • Step 332 Perform FreqTest: if PTDW Main Band is between Pulse Train Normal Range Table Stagger Min FreqBand and Max FreqBand, continue to Step 334 Add Positive Step Result, else continue to Step 338 Last PTDW?
  • Step 334 Add Positive Test result and continue to Step 336 Calculate location
  • Step 336 Calculate location, using Angle-of-arrival localization based on channel locations as described hereinbelow with reference to Figure 35
  • Step 338 If this is the last PTDW continue to Step 342 End, else continue to Step 340 Obtain next PTDW
  • Step 340 Obtain the next PTDW and continue to Step 324: PRI Type- Stagger?
  • the first PTDW PRI Type is Stagger and the Stagger level is 6 so the first pulse train is not a normal pulse train and results in a positive test result.
  • the third PTDW has a Stagger Level value 3 so it continues to Step 330 PRITest.
  • the third PTDW has a PTPRI value 870 is compared with Figure 34-C, PTNR Stagger first row which is relevant for stagger level 3 or below.
  • the PTNR Stagger first row Min PRI value is 400 and the Max PRI is 1200.
  • the third PTDW’s PTPRI value 870 is between these values so the third PTDW continues to Step 332 FreqTest.
  • the third PTDW has a Main Band value 4.
  • the PTNR Stagger first row Min Freq Band value is 8 and the Max Freq Band is 16.
  • the third PTDW’s Main Band value 4 is not between these values so the third pulse train does not result in a positive test result.
  • the third PTDW continues to Step 338 Last PTDW?
  • Pulse train tests preferably include location calculation in the event of a positive test result. Location is preferably assigned in accordance with the Talairach coordinate system. Location calculation is preferably performed on a pulse train’s main band. The calculation is based on a comparison of Single Channel Pulse Train (SCPT) ERPs in the strongest channels, typically 3 channels and can be extended depending on number of channels. SCPTs can include one or more single pulses. The test is performed on a pulse train’s main band.
  • SCPT Single Channel Pulse Train
  • Figure 35 A shows a flow diagram of location calculation includes the following steps:
  • Step 360 Start location calculation
  • Step 362 Obtain all SCPTs having a main band equal to the pulse train’s main band
  • Step 364 Find the 3 SCPTs from Step 362 with the strongest Max ERP and channel between 1 - 20
  • Step 366 Perform “weight” calculation on each SCPT where each channel weight is calculated by its ERP divided by the sum of all the channels ERP:
  • Step 368 Calculate Center of emission X, YZ coordinates by multiplying the Talairach coordinate values by the respective SCPT weights
  • Step 370 End
  • Figure 35B shows a PTDW of a pulse train which resulted in a positive test result.
  • the pulse train ’s main band is 13.
  • the pulse train includes 4 SCPTs.
  • Figure 35C shows the four SCPTDWs of the Figure 34 PTDW.
  • the channel 24 is not used for location calculation so the 3 strongest channels are channels 6, 8 and 16.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Veterinary Medicine (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Public Health (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Neurology (AREA)
  • Psychology (AREA)
  • Physiology (AREA)
  • Psychiatry (AREA)
  • Neurosurgery (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

La présente invention concerne un appareil de détection de tumeur de la tête servant à détecter des tumeurs de la tête à partir d'enregistrements d'EEG sur la base d'une extraction d'impulsions individuelles.
PCT/IL2021/051320 2020-12-02 2021-11-08 Appareil de détection de tumeur de la tête servant à détecter une tumeur de la tête et méthode associée WO2022118306A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
IL27916720 2020-12-02
IL279167 2020-12-02

Publications (1)

Publication Number Publication Date
WO2022118306A1 true WO2022118306A1 (fr) 2022-06-09

Family

ID=81854021

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IL2021/051320 WO2022118306A1 (fr) 2020-12-02 2021-11-08 Appareil de détection de tumeur de la tête servant à détecter une tumeur de la tête et méthode associée

Country Status (1)

Country Link
WO (1) WO2022118306A1 (fr)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005018448A1 (fr) * 2003-08-22 2005-03-03 Brainz Instruments Limited Analyse de capture d'eeg
WO2010115939A2 (fr) * 2009-04-07 2010-10-14 National University Of Ireland, Cork Méthode d'identification d'attaques en temps réel dans un signal d'électroencéphalogramme (eeg)
US20190110754A1 (en) 2017-10-17 2019-04-18 Satish Rao Machine learning based system for identifying and monitoring neurological disorders
US20190307345A1 (en) 2016-12-02 2019-10-10 Thomas Jefferson University Signal processing method for distinguishing and characterizing high-frequency oscillations

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005018448A1 (fr) * 2003-08-22 2005-03-03 Brainz Instruments Limited Analyse de capture d'eeg
WO2010115939A2 (fr) * 2009-04-07 2010-10-14 National University Of Ireland, Cork Méthode d'identification d'attaques en temps réel dans un signal d'électroencéphalogramme (eeg)
US20190307345A1 (en) 2016-12-02 2019-10-10 Thomas Jefferson University Signal processing method for distinguishing and characterizing high-frequency oscillations
US20190110754A1 (en) 2017-10-17 2019-04-18 Satish Rao Machine learning based system for identifying and monitoring neurological disorders

Non-Patent Citations (9)

* Cited by examiner, † Cited by third party
Title
ANKUSH SURKARPROF. NITIN AMBATKAR: "Primary Tumor detection with EEG Signals using Wavelet Transform and Neural network", INTERNATIONAL JOURNAL OF SCIENCE, ENGINEERING AND TECHNOLOGY RESEARCH, vol. 4, October 2015 (2015-10-01)
DR R. SUKANESHM. MURUGESAN: "Automated Detection of Brain Tumor in EEG Signals Using Artificial Neural Networks", INTERNATIONAL CONFERENCE ON ADVANCES IN COMPUTING, CONTROL, AND TELECOMMUNICATION TECHNOLOGIES, 2009
J. LATIKKAH. ESKOLA: "The Resistivity of Human Brain Tumours In Vivo", ANNALS OF BIOMEDICAL ENGINEERING, vol. 47, no. 3, March 2019 (2019-03-01), pages 706 - 713, XP036706715, DOI: 10.1007/s10439-018-02189-7
KUMAR ET AL.: "De-Interleaving and Identification of Pulsed Radar Signals Using ESM Receiver System", INTERNATIONAL JOURNAL OF ADVANCED RESEARCH IN ELECTRICAL, ELECTRONICS AND INSTRUMENTATION ENGINEERING, vol. 4, July 2015 (2015-07-01), pages 6243 - 6252
MASAKO OKAMOTOHARUKA DANKUNIKO SAKAMOTOKAZUHIRO TAKEOKOJI SHIMIZUSATORU KOHNOICHIRO ODASEIICHIRO ISOBETATEO SUZUKIKAORU KOHYAMA: "Three-dimensional probabilistic anatomical cranio-cerebral correlation via the international 10-20 system oriented for transcranial functional brain mapping", NEUROIMAGE, vol. 21, January 2004 (2004-01-01), pages 99 - 111
MUNTHER A.DAHLEHFADI N. KARAMEH: "Automated classification of EEG signals in brain tumor diagnostics", PROCEEDINGS OF THE AMERICAN CONTROL CONFERENCE CHICAGO, ILLINOIS, June 2000 (2000-06-01)
RAJAK BABLU ET AL: "Power Spectrum Density Analysis of EEG Signals in Spastic Cerebral Palsy Patients by Inducing r-TMS Therapy Design of Artificial Hand with Artificial Fingers View project Stroke Book View project POWER SPECTRAL ANALYSIS OF EEG AS A POTENTIAL MARKER IN THE DIAGNOSIS OF SPASTIC CEREBRAL PALSY CASES", ARTICLEININTERNATIONAL JOURNAL OF BIOMEDICAL ENGINEERING AND TECHNOLOGY INTERNATIONAL JOURNAL OF BIOMEDICAL ENGINEERING AND SCIENCE (IJBES), 1 January 2017 (2017-01-01), XP055885511, Retrieved from the Internet <URL:https://www.researchgate.net/profile/Meena-Gupta-6/publication/312053985_Power_Spectrum_Density_Analysis_of_EEG_Signals_in_Spastic_Cerebral_Palsy_Patients_by_Inducing_r-TMS_Therapy/links/58ca192492851c4b5e6ca231/Power-Spectrum-Density-Analysis-of-EEG-Signals-in-Spastic-Cerebral-Palsy-Patients-by-Ind> [retrieved on 20220131], DOI: 10.12691/jbet-4-1-2 *
SALAI SELVAM, V.S. SHENBAGADEVI: "Bispectral Analysis of Scalp Electroencephalograms: Quadratic Phase Coupling Phenomenon In Detecting Brain Tumor", AMERICAN JOURNAL OF APPLIED SCIENCES, vol. 10, no. 3, 2013, pages 294 - 306, ISSN: 1546-9239
V. SALAI SELVAMS. SHENBAGA DEVI: "Analysis of Spectral Features of EEG signal in Brain Tumor Condition", MEASUREMENT SCIENCE REVIEW, vol. 15, no. 4, 2015

Similar Documents

Publication Publication Date Title
Sharanreddy et al. EEG signal classification for epilepsy seizure detection using improved approximate entropy
US20150099962A1 (en) Methods and Systems To Identify Phase-Locked High-Frequency Oscillations In The Brain
Vakorin et al. Detecting mild traumatic brain injury using resting state magnetoencephalographic connectivity
US10588561B1 (en) Noninvasive system and method for mapping epileptic networks and surgical planning
JP2018023787A (ja) 波面のアノテーション
Migliorelli et al. Automated detection of epileptic ripples in MEG using beamformer-based virtual sensors
Duma et al. Altered spreading of neuronal avalanches in temporal lobe epilepsy relates to cognitive performance: A resting‐state hdEEG study
Abdi-Sargezeh et al. Incorporating uncertainty in data labeling into detection of brain interictal epileptiform discharges from EEG using weighted optimization
Chowdhury et al. Seizure activity classification based on bimodal Gaussian modeling of the gamma and theta band IMFs of EEG signals
US20210106247A1 (en) Methods and Apparatus for Detection and Imaging of Epileptogenicity from Scalp High-Frequency Oscillations
WO2022118306A1 (fr) Appareil de détection de tumeur de la tête servant à détecter une tumeur de la tête et méthode associée
Bakštein et al. Supervised segmentation of microelectrode recording artifacts using power spectral density
US20050273286A1 (en) Methods for identifying brain nuclei from micro-electrode signals
Ramović et al. Wavelet and Teager energy operator (TEO) for heart sound processing and identification
Liu et al. [Retracted] Automatic Detection of High‐Frequency Oscillations Based on an End‐to‐End Bi‐Branch Neural Network and Clinical Cross‐Validation
Lin et al. Brain Stroke Classification using a Microwave Transmission Line Approach
Abdi-Sargezeh et al. Mapping scalp to intracranial EEG using generative adversarial networks for automatically detecting interictal epileptiform discharges
Vetrikani et al. Diagnosis of epilepsy—A systematic review
CN113143292B (zh) 基于eeg和血清炎症因子分析的脑损伤标志物分析系统
Rajaei et al. Connectivity dynamics of interictal epileptiform activity
Wang et al. Diagnostic value of dynamic contrast-enhanced magnetic resonance imaging in Bell’s palsy
Sindhu et al. A review of magnetic resonance imaging and its clinical applications
Hu et al. A novel seizure diagnostic model based on generalized hurst exponent and extremely randomized trees
Wang et al. Lesion localization algorithm of high-frequency epileptic signal based on Teager energy operator
Grochowina et al. An application of wavelet transform for non-invasive evaluation of arteriovenous fistula state

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21810442

Country of ref document: EP

Kind code of ref document: A1

WPC Withdrawal of priority claims after completion of the technical preparations for international publication

Ref document number: 279167

Country of ref document: IL

Date of ref document: 20230516

Free format text: WITHDRAWN AFTER TECHNICAL PREPARATION FINISHED

NENP Non-entry into the national phase

Ref country code: DE