US-12622594-B2 - System for detection of breast cancer margin and method thereof

US12622594B2US 12622594 B2US12622594 B2US 12622594B2US-12622594-B2

Abstract

The disclosure discloses a multispectral imaging system and method for real time detection of breast cancer margin during a tumor resection surgery by combining terahertz and autofluorescence images. The system comprises terahertz and fluorescence modules in a housing having a sample holder to enable raster scan of the sample. The terahertz module has terahertz emitter antennas for generating terahertz radiation to the sample, and a terahertz detector to receive reflected terahertz signal from the sample. A fluorescence module with UV excitation LEDs induces fluorescence and a camera to receive emitted autofluorescence. A microcontroller connected to an electronic control unit is configured to display or to overlap and combine terahertz and autofluorescence images to determine breast cancer margin, via an AI module. The AI module is configured to perform classification of the feature set machine learning models to obtain classification with high with high sensitivity and specificity.

Inventors

Jyotirmayee DASH
Shaumik RAY
Balasubrahmanyam Pesala
Geethanjali RADHAKRISHNAN

Assignees

TERALUMEN SOLUTIONS PRIVATE LIMITED
ADIUVO DIAGNOSTICS PRIVATE LIMITED

Dates

Publication Date: 20260512
Application Date: 20231222
Priority Date: 20210623

Claims (8)

1 . A multispectral imaging system for real time detection of breast cancer margin during a tumor resection surgery by combining terahertz and autofluorescence images, comprising; a sample holder mounted on a stepper motor, adapted to hold and move a tissue sample in Z-axis and X-axis directions to enable raster scan of the sample; an optical controller, configured to accommodate at least one laser source to provide an optical input; a housing, configured to accommodate optical components, wherein the optical components include: a terahertz module having a terahertz emitter antenna adapted to generate terahertz radiation from the received optical input, one or more lenses to guide the generated terahertz radiation to the sample and a terahertz detector to receive reflected terahertz signal from the sample; a fluorescence module configured to generate and pass fluorescence radiation through the sample and to receive emitted autofluorescence, wherein the fluorescence module comprises: an imaging module configured to provide at least one excitation wavelength in infrared/ultraviolet range using one or more multiwavelength excitation LED to illuminate the sample to emit autofluorescence; an image acquisition module having one or more emission filters of predetermined wavelength adapted to filter the emitted autofluorescence from the sample; a camera to detect the filtered autofluorescence and generate an image based on autofluorescence; an electronic control unit connected to the optical controller to supply power, wherein the electronic control unit is configured to provide bias to the terahertz emitter antenna and form terahertz image with data acquisition unit; and a microcontroller connected to the electronic control unit, configured to display or to combine terahertz and autofluorescence images to determine breast cancer margin.
2 . The system as claimed in claim 1 , wherein a servo gear with a servo motor is employed for moving the emission filters.
3 . The system as claimed in claim 1 , wherein the microcontroller comprises a Graphical User Interface to display the processed images.
4 . The system as claimed in claim 1 , wherein at least one excitation wavelength is in the range of 300-500 nm.
5 . The system as claimed in claim 1 , wherein the predetermined wavelength of emission filters is 415 nm or 470 nm or 525 nm or 630 nm.
6 . A method of real time detection of breast cancer margin during a tumor resection surgery using a multispectral imaging system as claimed in claim 1 , the method comprising the steps of: providing the multispectral imaging system having a sample holder, an optical controller, a housing, an electronic control unit and a microcontroller, wherein the housing comprises a terahertz module with a terahertz emitter antenna, a terahertz detector antenna and a fluorescent module; providing and mounting a tissue sample onto the sample holder; generating optical input by at least one laser source in an optical controller; receiving the optical input in a terahertz module biased by an electronic control unit and guiding the generated terahertz radiation towards the sample; obtaining terahertz radiation from each pixel of the sample by terahertz detector antenna; generating fluorescence radiation in the fluorescence module and passing through the tissue sample to generate autofluorescence image; receiving reflected terahertz signal and autofluorescence response from the sample and capturing corresponding images; comparing and overlapping the obtained terahertz and autofluorescence images to corresponding stored histopathology images in the microcontroller; and segmenting cancer and noncancer regions, classifying and determining breast cancer margin by an artificial intelligence (AI) module.
7 . The method as claimed in claim 6 , wherein the method provides an improved breast cancer margin detection accuracy or specificity or sensitivity as compared to either the terahertz module or the fluorescence module alone.
8 . The method as claimed in claim 6 , wherein the AI module is configured for: capturing multispectral image data of an excised tissue sample; processing image data, resizing, analyzing and validating the preprocessed image data by overlapping image data with histopathology image; classifying the cancer and noncancer regions for each spectral image; combining spectral image validation data to form a feature set; and performing classification of the feature set by supervised vector machine or random forest machine learning model to obtain final classification with high sensitivity and specificity.

Description

BACKGROUND Technical Field The present disclosure relates to the detection of breast cancer margin and a system to detect the breast cancer margins. More particularly, the present disclosure relates to the detection of breast cancer margin after the removal of tumor or cancerous cells via surgery by employing Terahertz and Fluorescence radiations and spectroscopic analysis along with Artificial Intelligence based system. Description of the Related Art Breast cancer is one of the major causes of cancer related deaths across the world, especially in India and other developing nations. Breast cancer accounts for 14% of cancers in Indian women. Both in rural and urban India, the number of breast cancer cases have been seen to be on the rise. A 2018 report of Breast Cancer statistics recorded more than 1.6 lakh newly registered cases and 0.87 lakh reported deaths. Post cancer survival for women with breast cancer was reported 60% for Indian women, as compared to 80% in the U.S. The main reason for the higher mortality rate in India and other developing nations is the lack of proper infrastructure for effective diagnosis, delayed and incorrect cancer margin detection. The low survival rates can be addressed by accurate and timely detection of cancer margin which will also potentially reduce the requirement for repeat breast conservation surgeries. Breast conserving surgery (BCS) is an effective treatment for early-stage cancers as long as the margins of the respected tissue are free of disease according to consensus guidelines for patient management. 11 to 23% of patients undergo breast conservation surgery (BCS) in India instead of radical mastectomy (60 to 70% in western countries). BCS provides better cosmetic outcome, better psychological adjustment, less cost and better recovery. However, 20% of BCS cases fail due to inaccurate margin detection which means recurrence of cancer is more and there is a need of repeat surgery. However, 15% to 35% of patients undergo a second surgery since malignant cells are found close to or at the margins of the original resection specimen. The current gold standard technique for determination breast cancer margin is histopathology assessment which requires chemical staining and 2-7 days of time and analysis by a trained pathologist. Further to histopathology, two handheld probes, viz., the Margin Probe, developed by Dune Medical, Israel and ClearEdge, developed by LsBioPath, U.S.A. have been effectively employed for detection of breast cancer margin. Both these devices employ the same core method of determining tissue electrical properties, and thus malignancy, based on reflected wavelengths. However, it also has serious drawbacks of low sensitivity and specificity iKnife is another potential device based on rapid evaporative ionization mass spectrometry (REIMS) which has high sensitivity and specificity but is minimally invasive. At present, there are no diagnostic devices in India which can be employed for intraoperative assessment of breast cancer margin. The current gold standard technique is histopathology assessment which assesses the microscopic cellular structure of the tissue. These approaches are slower than other imaging techniques and usually more labor intensive and require surgical pathologists. In this process, the tissue will be fixed, processed, sectioned, stained with hematoxylin and eosin (H&E), and interpreted microscopically. Although the approach is accurate, it is also time-consuming and is completed over several days (5 to 7 days). Frozen sectioning is the most commonly used method for intraoperative margin assessment during surgery. Frozen section analysis consists of freezing small pieces of tissue, then sectioning, staining, and interpreting them under a microscope. The time between the tissue leaving the operating room (OR) and a microscopic diagnosis is more than one hour. The disadvantage of freezing tissue is that it generates significant artifacts, especially in fatty tissues, such as the breast. It can also be damaging such that tissue used for frozen sectioning may not be viable for the routine histopathologic margin assessment performed later. Frozen section interpretations are expensive and require additional technical staff to cut the specimens. These limitations result in a small fraction of the specimen margin being frozen and analyzed, leaving a large percentage of the tissue unassessed. Moreover, most of the smaller hospitals and multi-specialty clinics could not afford frozen section analysis. In these hospitals, patients have to wait for 3 to 7 days for histopathology results. Hence, a system/device is needed for intraoperative detection of the cancer margin rapidly with high specificity by conserving maximum normal tissue. Hence, there is a need of a device which can detect the breast cancer margin during surgical procedures of tumor removal and accurately detect breast cancer margin in non-invasive method with minimum skills. Prese