Artificial Intelligence Tool

14,000,000 Leading Edge Experts on the ideXlab platform

Scan Science and Technology

Contact Leading Edge Experts & Companies

Scan Science and Technology

Contact Leading Edge Experts & Companies

The Experts below are selected from a list of 17523 Experts worldwide ranked by ideXlab platform

Guzman Arenas Adolfo - One of the best experts on this subject based on the ideXlab platform.

  • Automatic mapping magnetic resonance images into multimedia database using SIFT
    IEEE Latin America Transactions, 2015
    Co-Authors: Reynoso MuÑoz, Jennifer Lynn, Cuevas Rasgado, Alma Delia, García Lamont Farid, Guzman Arenas Adolfo
    Abstract:

    This paper focuses on the representation of magnetic resonances of different parts of the human body, such as knees, spinal column, arms, elbows, etc., using ontologies. First, it maps the resonance images in a multimedia database. Then, automatically, using the SIFT pattern recognition algorithm, descriptors of the images stored in the database are extracted in order to recover useful data for the user; it uses the ontologies as an Artificial Intelligence Tool and, in consequence, reduces generation of useless data. Why do we think this is an interesting task? Because, if the user requires information about any topics or (s)he has some illness or needs to undergo magnetic resonance, this Tool will show him/her images and text to convey a better understanding, helping to obtain useful conclusions. Artificial Intelligence techniques are used, such as machine learning, knowledge representation, and pattern recognition. The ontological relations introduced here are based on the common representation of language, using definition dictionaries, Roget’s thesaurus, synonym dictionaries, and other resources. The system generates an output in the OM ontological language [1]. This language represents a structure where our system adds the data scanned by the SIFT algorithm. The tests have been made in Spanish; however, thanks to the portability of our system, it is possible to extend the method to any language.Proyecto UAEM 3454CHT/201

  • Automatic mapping magnetic resonance images into multimedia database using SIFT
    IEEE LATIN AMERICA TRANSACTIONS, 2015
    Co-Authors: Reynoso MuÑoz, Jennifer Lynn, Cuevas Rasgado, Alma Delia, García Lamont Farid, Guzman Arenas Adolfo
    Abstract:

    I. INTRODUCCIÓN STE proyecto de representación de la información a través de ontologías, analiza las imágenes de una base de datos multimedia, la cual contiene imágenes de tomografías, resonancia magnética de rodillas, brazos, columna vertebral (por citar algunas) con su descripción en texto y las ubica de manera automática en una ontología, que es nuestra base de conocimiento. Una ontología es un hipergrafo dirigido con vértices relacionados mediante aristas. En ella, un vértice representa un concepto o idea, mientras que un enlace representa la relación entre los vértices que une. Las características o propiedades de un concepto también se representan con aristas emanando del nodo correspondiente (Fig. 5). Nuestro sistema extrae imágenes de la base de datos multimedia, y coloca automáticamente cada imagen en el nodo correspondiente en la ontología, atendiendo a la categoría del objeto extraído. Esta ontología así enriquecida con imágenes es útil para consultas en comercio electrónico, aplicaciones 1 J. L. Reynoso, Universidad Autónoma del Estado de México, lynnreynoso@gmail.com. A. D. Cuevas, Universidad Autónoma del Estado de México, almadeliacuevas@gmail.com. F. García, Universidad Autónoma del Estado de México, fgarcial@uaemex.mx. A. Guzmán, Centro de Investigación en Computación del IPN , aguzman@ieee.org. médicas, rostros de criminales, marcas registradas, imágenes satelitales, etc. Para el reconocimiento de imágenes se usa el algoritmo SIFT (Scale Invariant Feature Transform) que extrae puntos clave que describen o modelan a los objetos en la escena [2]. Se construye un conjunto de entrenamiento que contiene los puntos clave extraídos de diferentes imágenes de objetos que se desean reconocer, en este caso, resonancias magnéticas de diferentes partes del cuerpo. En la fase de reconocimiento, a una imagen nueva se le extraen sus puntos clave y se comparan con los almacenados en el conjunto de entrenamiento para poder reconocer el objeto que aparece en la imagen. Es importante mencionar que los puntos clave son invariantes a la escala, rotación, pequeños cambios de iluminación y en la dirección de la vista, lo que hace que el reconocimiento sea robusto, hasta cierto punto. Además del algoritmo de reconocimiento de patrones y la carga automática de las imágenes de acuerdo al concepto al que pertenece, una interfaz del sistema permite a un usuario no sofisticado poder consultar las características de cada concept o y con ello su imagen. Está dirigido por ejemplo a usuarios no especializados en temas de medicina o a estudiantes de medicina que, por fines didácticos, pueden buscar información sobre un padecimiento a lo cual, una interfaz presenta información en texto estructurado como la glosa del concepto, palabras, idiomas, propiedades o características, imagen, nodos antecesores y sucesores en la ontología. La organización de este trabajo es la siguiente: En la sección I se presenta una introducción al tema, en la II se explica la principal idea que motivó nuestro desarrollo. En la sección III se presentan los conceptos básicos usados. La sección IV contiene los trabajos relacionados. En la sección V se presenta la metodología utilizada para el desarrollo del sistema. La sección VI contiene pruebas con los ejemplos de resonancia magnética de rodilla y resonancia magnética de columna lumbar, cráneo, brazo, pierna y mama. En la sección VII se muestran los resultados obtenidos de acuerdo a textos e imágenes analizadas. Las conclusiones aparecen en la sección VIII.This paper focuses on the representation of magnetic resonances of different parts of the human body, such as knees, spinal column, arms, elbows, etc., using ontologies. First, it maps the resonance im ages in a multimedia database. Then, automatically, using the SIFT pattern recognition algorithm, descriptors of the images stored in the database ar e extracted in order to recover useful data for the user; it use s the ontologies as an Artificial Intelligence Tool and, in consequen ce, reduces generation of useless data. Why do we think this is an interesting task? Because, if the user requires information abo ut any topics or (s)he has some illness or needs to undergo magnet ic resonance, this Tool will show him /her images and text to conve y a better understanding, helping t o obtain useful conclusions. Artificial Intelligence techniques are used, such as machine learning, knowledge representat ion, and pattern recognition. The ontological relations introduced here are based on the common representation of language , using definition dictionarie s, Roget’s thesaurus, synonym dictionaries, and other resources The system generates an output in the OM ontological language [1]. This language represents a structure where our system adds the data scanned by the SIFT algorithm. The tests have been made in Spanish; however, thanks to the portability of our system, it is possible to extend the method to any language.Sistema Nacional de Investigadores SNI, El Consejo Nacional de Ciencia y Tecnología CONACYT del gobierno Mexicano, La Secretaría de Investigación y Posgrado del Instituto Politécnico Nacional SIP-IPN. Universidad Autónoma del Estado de México, Centro Universitario Texcoco La Secretaría de Investigación y Estudios Avanzados SIEA. El proyecto: 3454CHT/201

Reynoso MuÑoz, Jennifer Lynn - One of the best experts on this subject based on the ideXlab platform.

  • Automatic mapping magnetic resonance images into multimedia database using SIFT
    IEEE Latin America Transactions, 2015
    Co-Authors: Reynoso MuÑoz, Jennifer Lynn, Cuevas Rasgado, Alma Delia, García Lamont Farid, Guzman Arenas Adolfo
    Abstract:

    This paper focuses on the representation of magnetic resonances of different parts of the human body, such as knees, spinal column, arms, elbows, etc., using ontologies. First, it maps the resonance images in a multimedia database. Then, automatically, using the SIFT pattern recognition algorithm, descriptors of the images stored in the database are extracted in order to recover useful data for the user; it uses the ontologies as an Artificial Intelligence Tool and, in consequence, reduces generation of useless data. Why do we think this is an interesting task? Because, if the user requires information about any topics or (s)he has some illness or needs to undergo magnetic resonance, this Tool will show him/her images and text to convey a better understanding, helping to obtain useful conclusions. Artificial Intelligence techniques are used, such as machine learning, knowledge representation, and pattern recognition. The ontological relations introduced here are based on the common representation of language, using definition dictionaries, Roget’s thesaurus, synonym dictionaries, and other resources. The system generates an output in the OM ontological language [1]. This language represents a structure where our system adds the data scanned by the SIFT algorithm. The tests have been made in Spanish; however, thanks to the portability of our system, it is possible to extend the method to any language.Proyecto UAEM 3454CHT/201

  • Automatic mapping magnetic resonance images into multimedia database using SIFT
    IEEE LATIN AMERICA TRANSACTIONS, 2015
    Co-Authors: Reynoso MuÑoz, Jennifer Lynn, Cuevas Rasgado, Alma Delia, García Lamont Farid, Guzman Arenas Adolfo
    Abstract:

    I. INTRODUCCIÓN STE proyecto de representación de la información a través de ontologías, analiza las imágenes de una base de datos multimedia, la cual contiene imágenes de tomografías, resonancia magnética de rodillas, brazos, columna vertebral (por citar algunas) con su descripción en texto y las ubica de manera automática en una ontología, que es nuestra base de conocimiento. Una ontología es un hipergrafo dirigido con vértices relacionados mediante aristas. En ella, un vértice representa un concepto o idea, mientras que un enlace representa la relación entre los vértices que une. Las características o propiedades de un concepto también se representan con aristas emanando del nodo correspondiente (Fig. 5). Nuestro sistema extrae imágenes de la base de datos multimedia, y coloca automáticamente cada imagen en el nodo correspondiente en la ontología, atendiendo a la categoría del objeto extraído. Esta ontología así enriquecida con imágenes es útil para consultas en comercio electrónico, aplicaciones 1 J. L. Reynoso, Universidad Autónoma del Estado de México, lynnreynoso@gmail.com. A. D. Cuevas, Universidad Autónoma del Estado de México, almadeliacuevas@gmail.com. F. García, Universidad Autónoma del Estado de México, fgarcial@uaemex.mx. A. Guzmán, Centro de Investigación en Computación del IPN , aguzman@ieee.org. médicas, rostros de criminales, marcas registradas, imágenes satelitales, etc. Para el reconocimiento de imágenes se usa el algoritmo SIFT (Scale Invariant Feature Transform) que extrae puntos clave que describen o modelan a los objetos en la escena [2]. Se construye un conjunto de entrenamiento que contiene los puntos clave extraídos de diferentes imágenes de objetos que se desean reconocer, en este caso, resonancias magnéticas de diferentes partes del cuerpo. En la fase de reconocimiento, a una imagen nueva se le extraen sus puntos clave y se comparan con los almacenados en el conjunto de entrenamiento para poder reconocer el objeto que aparece en la imagen. Es importante mencionar que los puntos clave son invariantes a la escala, rotación, pequeños cambios de iluminación y en la dirección de la vista, lo que hace que el reconocimiento sea robusto, hasta cierto punto. Además del algoritmo de reconocimiento de patrones y la carga automática de las imágenes de acuerdo al concepto al que pertenece, una interfaz del sistema permite a un usuario no sofisticado poder consultar las características de cada concept o y con ello su imagen. Está dirigido por ejemplo a usuarios no especializados en temas de medicina o a estudiantes de medicina que, por fines didácticos, pueden buscar información sobre un padecimiento a lo cual, una interfaz presenta información en texto estructurado como la glosa del concepto, palabras, idiomas, propiedades o características, imagen, nodos antecesores y sucesores en la ontología. La organización de este trabajo es la siguiente: En la sección I se presenta una introducción al tema, en la II se explica la principal idea que motivó nuestro desarrollo. En la sección III se presentan los conceptos básicos usados. La sección IV contiene los trabajos relacionados. En la sección V se presenta la metodología utilizada para el desarrollo del sistema. La sección VI contiene pruebas con los ejemplos de resonancia magnética de rodilla y resonancia magnética de columna lumbar, cráneo, brazo, pierna y mama. En la sección VII se muestran los resultados obtenidos de acuerdo a textos e imágenes analizadas. Las conclusiones aparecen en la sección VIII.This paper focuses on the representation of magnetic resonances of different parts of the human body, such as knees, spinal column, arms, elbows, etc., using ontologies. First, it maps the resonance im ages in a multimedia database. Then, automatically, using the SIFT pattern recognition algorithm, descriptors of the images stored in the database ar e extracted in order to recover useful data for the user; it use s the ontologies as an Artificial Intelligence Tool and, in consequen ce, reduces generation of useless data. Why do we think this is an interesting task? Because, if the user requires information abo ut any topics or (s)he has some illness or needs to undergo magnet ic resonance, this Tool will show him /her images and text to conve y a better understanding, helping t o obtain useful conclusions. Artificial Intelligence techniques are used, such as machine learning, knowledge representat ion, and pattern recognition. The ontological relations introduced here are based on the common representation of language , using definition dictionarie s, Roget’s thesaurus, synonym dictionaries, and other resources The system generates an output in the OM ontological language [1]. This language represents a structure where our system adds the data scanned by the SIFT algorithm. The tests have been made in Spanish; however, thanks to the portability of our system, it is possible to extend the method to any language.Sistema Nacional de Investigadores SNI, El Consejo Nacional de Ciencia y Tecnología CONACYT del gobierno Mexicano, La Secretaría de Investigación y Posgrado del Instituto Politécnico Nacional SIP-IPN. Universidad Autónoma del Estado de México, Centro Universitario Texcoco La Secretaría de Investigación y Estudios Avanzados SIEA. El proyecto: 3454CHT/201

Benjamin Moxley-wyles - One of the best experts on this subject based on the ideXlab platform.

  • Artificial Intelligence for advance requesting of immunohistochemistry in diagnostically uncertain prostate biopsies
    Modern Pathology, 2021
    Co-Authors: Andrea Chatrian, Korsuk Sirinukunwattana, Nasullah Khalid Alham, Lisa Browning, S Malacrino, M Haghighat, A Aberdeen, A Monks, Richard T. Colling, Benjamin Moxley-wyles
    Abstract:

    The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel Artificial Intelligence Tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide-level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by threefold cross-validation. Validation was conducted on a separate validation dataset of 222 images. Non IHC-requested cases were diagnosed in 17.9 min on average, while IHC-requested cases took 33.4 min over multiple reporting sessions. We estimated 11 min could be saved on average per case by automated IHC requesting, by removing duplication of effort. The Tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI Tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.

Korsuk Sirinukunwattana - One of the best experts on this subject based on the ideXlab platform.

  • Artificial Intelligence for advance requesting of immunohistochemistry in diagnostically uncertain prostate biopsies
    Modern Pathology, 2021
    Co-Authors: Andrea Chatrian, Korsuk Sirinukunwattana, Nasullah Khalid Alham, Lisa Browning, S Malacrino, M Haghighat, A Aberdeen, A Monks, Richard T. Colling, Benjamin Moxley-wyles
    Abstract:

    The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel Artificial Intelligence Tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide-level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by threefold cross-validation. Validation was conducted on a separate validation dataset of 222 images. Non IHC-requested cases were diagnosed in 17.9 min on average, while IHC-requested cases took 33.4 min over multiple reporting sessions. We estimated 11 min could be saved on average per case by automated IHC requesting, by removing duplication of effort. The Tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI Tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.

  • the potential of Artificial Intelligence to detect lymphovascular invasion in testicular cancer
    Cancers, 2021
    Co-Authors: Abhisek Ghosh, Korsuk Sirinukunwattana, Nasullah Khalid Alham, Lisa Browning, Richard Colling, Andrew Protheroe, Emily Protheroe, Stephanie R Jones
    Abstract:

    Testicular cancer is the most common cancer in men aged from 15 to 34 years. Lymphovascular invasion refers to the presence of tumours within endothelial-lined lymphatic or vascular channels, and has been shown to have prognostic significance in testicular germ cell tumours. In non-seminomatous tumours, lymphovascular invasion is the most powerful prognostic factor for stage 1 disease. For the pathologist, searching multiple slides for lymphovascular invasion can be highly time-consuming. The aim of this retrospective study was to develop and assess an Artificial Intelligence algorithm that can identify areas suspicious for lymphovascular invasion in histological digital whole slide images. Areas of possible lymphovascular invasion were annotated in a total of 184 whole slide images of haematoxylin and eosin (H&E) stained tissue from 19 patients with testicular germ cell tumours, including a mixture of seminoma and non-seminomatous cases. Following consensus review by specialist uropathologists, we trained a deep learning classifier for automatic segmentation of areas suspicious for lymphovascular invasion. The classifier identified 34 areas within a validation set of 118 whole slide images from 10 patients, each of which was reviewed by three expert pathologists to form a majority consensus. The precision was 0.68 for areas which were considered to be appropriate to flag, and 0.56 for areas considered to be definite lymphovascular invasion. An Artificial Intelligence Tool which highlights areas of possible lymphovascular invasion to reporting pathologists, who then make a final judgement on its presence or absence, has been demonstrated as feasible in this proof-of-concept study. Further development is required before clinical deployment.

  • Artificial Intelligence for advance requesting of immunohistochemistry in diagnostically uncertain prostate biopsies
    medRxiv, 2021
    Co-Authors: Andrea Chatrian, Korsuk Sirinukunwattana, Nasullah Khalid Alham, R Colling, L Browning, S Malacrino, M Haghighat, A Aberdeen, A Monks, B Moxleywyles
    Abstract:

    The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel Artificial Intelligence Tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by 3-fold cross- validation. Validation was conducted on a separate validation dataset of 212 images. Non IHC-requested cases were diagnosed in 17.9 minutes on average, while IHC-requested cases took 33.4 minutes over multiple reporting sessions. We estimated 11 minutes could be saved on average per case by automated IHC requesting, by removing duplication of effort. The Tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI Tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.

Nasullah Khalid Alham - One of the best experts on this subject based on the ideXlab platform.

  • Artificial Intelligence for advance requesting of immunohistochemistry in diagnostically uncertain prostate biopsies
    Modern Pathology, 2021
    Co-Authors: Andrea Chatrian, Korsuk Sirinukunwattana, Nasullah Khalid Alham, Lisa Browning, S Malacrino, M Haghighat, A Aberdeen, A Monks, Richard T. Colling, Benjamin Moxley-wyles
    Abstract:

    The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel Artificial Intelligence Tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide-level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by threefold cross-validation. Validation was conducted on a separate validation dataset of 222 images. Non IHC-requested cases were diagnosed in 17.9 min on average, while IHC-requested cases took 33.4 min over multiple reporting sessions. We estimated 11 min could be saved on average per case by automated IHC requesting, by removing duplication of effort. The Tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI Tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.

  • the potential of Artificial Intelligence to detect lymphovascular invasion in testicular cancer
    Cancers, 2021
    Co-Authors: Abhisek Ghosh, Korsuk Sirinukunwattana, Nasullah Khalid Alham, Lisa Browning, Richard Colling, Andrew Protheroe, Emily Protheroe, Stephanie R Jones
    Abstract:

    Testicular cancer is the most common cancer in men aged from 15 to 34 years. Lymphovascular invasion refers to the presence of tumours within endothelial-lined lymphatic or vascular channels, and has been shown to have prognostic significance in testicular germ cell tumours. In non-seminomatous tumours, lymphovascular invasion is the most powerful prognostic factor for stage 1 disease. For the pathologist, searching multiple slides for lymphovascular invasion can be highly time-consuming. The aim of this retrospective study was to develop and assess an Artificial Intelligence algorithm that can identify areas suspicious for lymphovascular invasion in histological digital whole slide images. Areas of possible lymphovascular invasion were annotated in a total of 184 whole slide images of haematoxylin and eosin (H&E) stained tissue from 19 patients with testicular germ cell tumours, including a mixture of seminoma and non-seminomatous cases. Following consensus review by specialist uropathologists, we trained a deep learning classifier for automatic segmentation of areas suspicious for lymphovascular invasion. The classifier identified 34 areas within a validation set of 118 whole slide images from 10 patients, each of which was reviewed by three expert pathologists to form a majority consensus. The precision was 0.68 for areas which were considered to be appropriate to flag, and 0.56 for areas considered to be definite lymphovascular invasion. An Artificial Intelligence Tool which highlights areas of possible lymphovascular invasion to reporting pathologists, who then make a final judgement on its presence or absence, has been demonstrated as feasible in this proof-of-concept study. Further development is required before clinical deployment.

  • Artificial Intelligence for advance requesting of immunohistochemistry in diagnostically uncertain prostate biopsies
    medRxiv, 2021
    Co-Authors: Andrea Chatrian, Korsuk Sirinukunwattana, Nasullah Khalid Alham, R Colling, L Browning, S Malacrino, M Haghighat, A Aberdeen, A Monks, B Moxleywyles
    Abstract:

    The use of immunohistochemistry in the reporting of prostate biopsies is an important adjunct when the diagnosis is not definite on haematoxylin and eosin (H&E) morphology alone. The process is however inherently inefficient with delays while waiting for pathologist review to make the request and duplicated effort reviewing a case more than once. In this study, we aimed to capture the workflow implications of immunohistochemistry requests and demonstrate a novel Artificial Intelligence Tool to identify cases in which immunohistochemistry (IHC) is required and generate an automated request. We conducted audits of the workflow for prostate biopsies in order to understand the potential implications of automated immunohistochemistry requesting and collected prospective cases to train a deep neural network algorithm to detect tissue regions that presented ambiguous morphology on whole slide images. These ambiguous foci were selected on the basis of the pathologist requesting immunohistochemistry to aid diagnosis. A gradient boosted trees classifier was then used to make a slide level prediction based on the outputs of the neural network prediction. The algorithm was trained on annotations of 219 immunohistochemistry-requested and 80 control images, and tested by 3-fold cross- validation. Validation was conducted on a separate validation dataset of 212 images. Non IHC-requested cases were diagnosed in 17.9 minutes on average, while IHC-requested cases took 33.4 minutes over multiple reporting sessions. We estimated 11 minutes could be saved on average per case by automated IHC requesting, by removing duplication of effort. The Tool attained 99% accuracy and 0.99 Area Under the Curve (AUC) on the test data. In the validation, the average agreement with pathologists was 0.81, with a mean AUC of 0.80. We demonstrate the proof-of-principle that an AI Tool making automated immunohistochemistry requests could create a significantly leaner workflow and result in pathologist time savings.

Related terms

Artificial Intelligence Tool - 15 days free trial to Access Experts & Abstracts