Artificial intelligence for pancreatic cancer detection: Recent development and future direction

Machine Learning is the study of algorithms and application of artificial intelligence. Artificial Intelligence is said to be the superset of machine learning. It aims to develop those programs which can learn and improves it upon increasing experience. It is designed to learn by itself. The aim is to detect pancreatic cancer using machine learning approach. Pancreas is responsible for secreting insulin which helps to control the blood glucose level in the human body. The paper aims to detect pancreatic cancer with the help of machine learning. The tumor is detected using image processing and is to be detected at the premature stage so that proper medication and treatment can be provided to increase the survival rate of the patient. The MRI image of pancreas obtained after MRI scan will be preprocessed and its noise is removed. The segmentation of MRI images will be performed using FCM algorithm. The tumor present in the image will be detected with the help of morphological process and multi clustering model. After Segmentation the image will be divided into various regions. With the help of the hybrid technique the primary and secondary regions are compressed and are used for telemedicine application. DWT is used for DE noising the image. GLCM features are extracted. The image then compared with the database images of pancreatic tumors and is classified as abnormal and normal with the help of BPN based classifier. The image is classified into abnormal and normal. The malignant image is considered as abnormal. The abnormal image is then segmented using SFCM and tumor part is clustered. After clustering the tumor part validation about the presence of pancreatic cancer is given.

Download Full-text

Systems Biology Approaches to Pancreatic Cancer Detection, Prevention and Treatment

Current Pharmaceutical Design ◽

10.2174/138161282001140113124643 ◽

2014 ◽

Vol 20 (1) ◽

pp. 73-80 ◽

Cited By ~ 5

Author(s):

Osama Alian ◽

Philip Philip ◽

Fazlul Sarkar ◽

Asfar Azmi

Keyword(s):

Pancreatic Cancer ◽

Systems Biology ◽

Cancer Detection ◽

Prevention And Treatment

Download Full-text

Tumor microenvironment and the role of artificial intelligence in breast cancer detection and prognosis

American Journal Of Pathology ◽

10.1016/j.ajpath.2021.01.014 ◽

2021 ◽

Author(s):

Kathryn Malherbe.

Keyword(s):

Breast Cancer ◽

Artificial Intelligence ◽

Tumor Microenvironment ◽

Cancer Detection ◽

Breast Cancer Detection

Download Full-text

Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study

European Radiology ◽

10.1007/s00330-021-07992-w ◽

2021 ◽

Author(s):

Suzanne L. van Winkel ◽

Alejandro Rodríguez-Ruiz ◽

Linda Appelman ◽

Albert Gubern-Mérida ◽

Nico Karssemeijer ◽

...

Keyword(s):

Breast Cancer ◽

Artificial Intelligence ◽

Cancer Detection ◽

Support System ◽

Reading Time ◽

Digital Breast Tomosynthesis ◽

Breast Tomosynthesis ◽

Screening Programs ◽

The Impact

Abstract Objectives Digital breast tomosynthesis (DBT) increases sensitivity of mammography and is increasingly implemented in breast cancer screening. However, the large volume of images increases the risk of reading errors and reading time. This study aims to investigate whether the accuracy of breast radiologists reading wide-angle DBT increases with the aid of an artificial intelligence (AI) support system. Also, the impact on reading time was assessed and the stand-alone performance of the AI system in the detection of malignancies was compared to the average radiologist. Methods A multi-reader multi-case study was performed with 240 bilateral DBT exams (71 breasts with cancer lesions, 70 breasts with benign findings, 339 normal breasts). Exams were interpreted by 18 radiologists, with and without AI support, providing cancer suspicion scores per breast. Using AI support, radiologists were shown examination-based and region-based cancer likelihood scores. Area under the receiver operating characteristic curve (AUC) and reading time per exam were compared between reading conditions using mixed-models analysis of variance. Results On average, the AUC was higher using AI support (0.863 vs 0.833; p = 0.0025). Using AI support, reading time per DBT exam was reduced (p < 0.001) from 41 (95% CI = 39–42 s) to 36 s (95% CI = 35– 37 s). The AUC of the stand-alone AI system was non-inferior to the AUC of the average radiologist (+0.007, p = 0.8115). Conclusions Radiologists improved their cancer detection and reduced reading time when evaluating DBT examinations using an AI reading support system. Key Points • Radiologists improved their cancer detection accuracy in digital breast tomosynthesis (DBT) when using an AI system for support, while simultaneously reducing reading time. • The stand-alone breast cancer detection performance of an AI system is non-inferior to the average performance of radiologists for reading digital breast tomosynthesis exams. • The use of an AI support system could make advanced and more reliable imaging techniques more accessible and could allow for more cost-effective breast screening programs with DBT.

Download Full-text

Progress and potential of exosome analysis for early pancreatic cancer detection

Expert Review of Molecular Diagnostics ◽

10.1080/14737159.2016.1187563 ◽

2016 ◽

Vol 16 (7) ◽

pp. 757-767 ◽

Cited By ~ 5

Author(s):

Ulrike Erb ◽

Margot Zöller

Keyword(s):

Pancreatic Cancer ◽

Cancer Detection

Download Full-text

APPLICATION OF ARTIFICIAL INTELLIGENCE TECHNIQUES IN MS. PAC-MAN GAME: A REVIEW

International Journal of Creative Future and Heritage (TENIAT) ◽

10.47252/teniat.v3i1.251 ◽

2015 ◽

Vol 3 (1) ◽

Author(s):

Tse Guan Tan ◽

Jason Teo

Keyword(s):

Artificial Intelligence ◽

Content Analysis ◽

Video Games ◽

Analysis Method ◽

Engineering Economics ◽

Game Ai ◽

Content Analysis Method ◽

Wide Range ◽

Computer Based ◽

Future Direction

AbstrakTeknik Kecerdasan Buatan (AI) berjaya digunakan dan diaplikasikan dalam pelbagai bidang, termasukpembuatan, kejuruteraan, ekonomi, perubatan dan ketenteraan. Kebelakangan ini, terdapat minat yangsemakin meningkat dalam Permainan Kecerdasan Buatan atau permainan AI. Permainan AI merujukkepada teknik yang diaplikasikan dalam permainan komputer dan video seperti pembelajaran, pathfinding,perancangan, dan lain-lain bagi mewujudkan tingkah laku pintar dan autonomi kepada karakter dalampermainan. Objektif utama kajian ini adalah untuk mengemukakan beberapa teknik yang biasa digunakandalam merekabentuk dan mengawal karakter berasaskan komputer untuk permainan Ms Pac-Man antaratahun 2005-2012. Ms Pac-Man adalah salah satu permainan yang digunakan dalam siri pertandinganpermainan diperingkat antarabangsa sebagai penanda aras untuk perbandingan pengawal autonomi.Kaedah analisis kandungan yang menyeluruh dijalankan secara ulasan dan sorotan literatur secara kritikal.Dapatan kajian menunjukkan bahawa, walaupun terdapat berbagai teknik, limitasi utama dalam kajianterdahulu untuk mewujudkan karakter permaianan Pac Man adalah kekurangan Generalization Capabilitydalam kepelbagaian karakter permainan. Hasil kajian ini akan dapat digunakan oleh penyelidik untukmeningkatkan keupayaan Generalization AI karakter permainan dalam Pasaran Permainan KecerdasanBuatan. Abstract Artificial Intelligence (AI) techniques are successfully used and applied in a wide range of areas, includingmanufacturing, engineering, economics, medicine and military. In recent years, there has been anincreasing interest in Game Artificial Intelligence or Game AI. Game AI refers to techniques applied incomputer and video games such as learning, pathfinding, planning, and many others for creating intelligentand autonomous behaviour to the characters in games. The main objective of this paper is to highlightseveral most common of the AI techniques for designing and controlling the computer-based charactersto play Ms. Pac-Man game between years 2005-2012. The Ms. Pac-Man is one of the games that used asbenchmark for comparison of autonomous controllers in a series of international Game AI competitions.An extensive content analysis method was conducted through critical review on previous literature relatedto the field. Findings highlight, although there was various and unique techniques available, the majorlimitation of previous studies for creating the Ms. Pac-Man game characters is a lack of generalizationcapability across different game characters. The findings could provide the future direction for researchersto improve the Generalization A.I capability of game characters in the Game Artificial Intelligence market.

Download Full-text

Development of extracellular vesicles-based classifier for detection of early-stage bladder, ovarian, and pancreatic cancer.

Journal of Clinical Oncology ◽

10.1200/jco.2021.39.15_suppl.3048 ◽

2021 ◽

Vol 39 (15_suppl) ◽

pp. 3048-3048

Author(s):

Juan Pablo Hinestrosa ◽

Razelle Kurzrock ◽

Jean Lewis ◽

Nick Schork ◽

Ashish M. Kamat ◽

...

Keyword(s):

Pancreatic Cancer ◽

Confidence Interval ◽

Extracellular Vesicles ◽

Metastatic Disease ◽

Cancer Detection ◽

Early Stage ◽

Expression Patterns ◽

Stepwise Logistic Regression ◽

Liquid Biopsies ◽

Cancer Early Detection

3048 Background: Many cancers are lethal because they present with metastatic disease. Because localized/resectable tumors produce vague symptoms, diagnosis is delayed. In pancreatic cancer, only ̃10% of patients survive five years, and it will soon become the second leading cause of cancer-related deaths in the USA. For patients with metastatic disease, the 2- and 5-year survival is < 10% and ̃3%, respectively. For the few patients with local disease, 5-year survival is ̃40%. Many other cancers have comparable differences between early- and late-stage disease. It is apparent a diagnostic assay for early-stage cancers would transform the field by minimizing the need for aggressive surgeries and other harsh interventions, and by its potential to increase survival. Identifying cancer-specific aberrations in blood-based “liquid” biopsies offers a prospect for a non-invasive cancer detection tool. In the bloodstream, there are extracellular vesicles (EVs) with cargoes including membrane and cytosolic proteins, as well as RNA and lipids derived from their parent cells. Methods: We used an alternating current electrokinetics (ACE) microarray to isolate EVs from the plasma of stage I and II bladder (N = 48), ovarian (N = 42), and pancreatic cancer patients (N = 44), and healthy volunteers (N = 110). EVs were analyzed using multiplex protein immunoassays for 54 cancer-related proteins. EV protein expression patterns were analyzed using stepwise logistic regression followed by a split between training and test sets (67%/33% respectively). This process enabled biomarker selection and generation of a classifier to discriminate between cancer and healthy donors. Results: The EV protein-based classifier had an overall area under curve (AUC) of 0.95 with a sensitivity of 71.2% (69.4% – 73.0%, at 95% confidence interval) at > 99% specificity. The classifier’s performance for the pancreatic cancer cohort was very strong, with overall sensitivity of 95.7% (94.6% – 96.9%, at 95% confidence interval) at > 99% specificity. Conclusions: EV-associated proteins may enable early cancer detection where surgical resection is most likely to improve outcomes. The classifier’s performance for the initial three cancers studied showed encouraging results. Future efforts will include examining additional cancer types and evaluating the classifier performance using samples from donors with related benign conditions with the aim of a pan-cancer early detection assay.

Download Full-text