An Article on Pituitary Adenoma detection through CNN and ANN

Akash Pandey

Assistant Professor

CS & IT Dept.

Kalinga University

akash.pandey@kalingauniversity.ac.in

Introduction:

Common noncancerous tumors called pituitary adenomas grow in the pituitary gland, a tiny organ the size of a pea that is situated near the base of the brain. Although the majority of pituitary adenomas are benign, timely and correct identification is necessary due to their potential for development and impact on hormone control. The identification and characterization of these tumors is greatly aided by medical imaging, especially MRI and CT scans. Automating the diagnosis of pituitary adenomas has showed promise in recent years with the introduction of machine learning techniques, particularly Convolutional Neural Networks (CNN) and Artificial Neural Networks (ANN). These sophisticated computational algorithms can evaluate medical images, extract pertinent information, and produce very accurate predictions.

Keywords: MRI, CT, CNN, ANN, Pituitary gland.

Importance of Early Detection:

Pituitary adenomas should be detected as soon as possible for a number of reasons, including: Better treatment outcomes due to timely action following early identification of the condition.

Prevention of Complications: Early identification reduces the risk of complications brought on by hormone abnormalities and tumor development.

Healthcare providers can create individualized and successful treatment regimens with the help of improved patient management and effective detection.

Role of Machine Learning in Pituitary Adenoma Detection:

Pituitary adenoma diagnosis can be done automatically and with data using machine learning models, especially CNN and ANN. These models are capable of deciphering intricate patterns found in medical photos and offering insightful information for precise diagnosis. The goal of incorporating machine learning technologies into medical procedures is to improve the accuracy and efficiency of diagnostic procedures.

Convolutional Neural Networks (CNN) for Image Analysis:

CNNs are perfect for evaluating medical imaging data because they are well-suited for image-related tasks. CNNs are able to automatically recognize complex elements in photos, such as nuances that may indicate pituitary adenomas, thanks to their hierarchical nature. CNNs perform exceptionally well at capturing spatial relationships within the images because to convolutional layers and pooling processes.

Artificial Neural Networks (ANN) for Classification:

In the domain of pituitary adenoma diagnosis, ANNs—which are especially made for structured data—supplement CNNs’ capabilities. ANNs are capable of creating binary classifications, or identifying between images with and without adenomas, using features that have been retrieved from images. Artificial neural networks (ANNs) are useful in decision-making because of their capacity to learn from intricate relationships seen in data.

Methodology:

Data Acquisition:

A tagged medical picture dataset is gathered that includes both positive and negative cases of pituitary adenomas.

Preprocessing:

To concentrate on the pituitary gland, images are normalized and pertinent areas of interest (ROIs) are extracted.

Model Development:

CNNs are used in feature extraction to extract complex patterns from the images. An ANN receives the extracted features in order to perform binary classification.

Training:

Using the labeled dataset, the model is trained with the goal of maximizing accuracy and generalization.

Validation and Testing:

The model is tested for real-world scenarios and validated on an independent dataset to evaluate its performance.

Conclusion:

The combination of CNN and ANN for the diagnosis of pituitary adenoma has enormous potential to transform the diagnostic procedure. Healthcare practitioners can identify pituitary adenomas more quickly, accurately, and efficiently by utilizing machine learning, which will ultimately improve patient outcomes and healthcare delivery. These technologies are projected to play an increasingly important role in diagnostic radiography as they develop, supporting the continuous advancement of precision medicine.

References:

1. Park JG, Lee C (2009) Skull stripping based on region growingNeuroimage47:1394–1407
2. Pandey and S. K. Shrivastava, “A Survey Paper on Calcaneus Bone Tumor Detection Using different Improved Canny Edge Detector,” 2018 IEEE International Conference on System, Computation, Automation and Networking (ICSCA), Pondicherry, India, 2018, pp. 1-5, doi: 10.1109/ICSCAN.2018.8541194.

3. Khan MA, Lali IU, Rehman A, Ishaq M, Sharif M, Saba T et al(2019)Braintumordetectionandclassification:Aframeworkofmarker-basedwatershedalgorithmandmultilevelMicroscResTech82:909–922.

4. RazaM,SharifM,YasminM,MasoodS,MohsinS(2012)Brainimagerepresentationandrendering:asurvey.ResJApplSciEngTechnol4:3274–3282

5. Watson C, Kirkcaldie M, Paxinos G (2010) The brain: an intro-ductiontofunctionalAcademicPress,NewYork

6. (2015).https://en.wikipedia.org/wiki/Brain_size.Accessed19Oct2019

7. DubinMW(2013)HowWiley,NewYork

8. Koziol LF, Budding DE, Chidekel D (2012) From movement tothought: executive function, embodied cognition, and the cerebel-Cerebellum11:505–525

9. Knierim J (1997) Neuroscience Online Chapter 5: Cerebellum.TheUniversityofTexasHealthScienceCenter,Houston.

10. Nuñez MA, Miranda JCF, de Oliveira E, Rubino PA, VoscoboinikS,RecaldeRetal(2019)Brainstemanatomyandsurgi-cal approaches. Comprehensive overview of modern surgicalapproaches to intrinsic brain tumors. Elsevier, Amsterdam, pp53–105

11. DeAngelisLM(2001)Braintumors.NEnglJMed344:114–123

12. Amin J, Sharif M, Raza M, Saba T, Sial R, Shad SA (2020) Braintumordetection:alongshort-termmemory(LSTM)-basedlearn-ingmodel.NeuralComputAppl32:15965–15973

13. Sajjad S, Hanan Abdullah A, Sharif M, Mohsin S (2014) Psy-chotherapythroughvideogametotargetillnessrelatedproblem-atic behaviors of children with brain tumor. Curr Med Imaging10:62–72

14. YasminM,SharifM,MasoodS,RazaM,MohsinS(2012)Brainimagereconstruction:ashortsurvey.WorldApplSciJ19:52–62

15. AminJ,SharifM,RazaM,YasminM(2018)Detectionofbraintumorbasedonfeaturesfusionandmachinelearning.JAmbientIntellHuman Comput:1–17

16. AminJ,SharifM,YasminM,FernandesSL(2020)Adistinctiveapproach in brain tumor detection and classification using MRI.PatternRecognLett139:118–127

17. Saba T, Mohamed AS, El-Affendi M, Amin J, Sharif M (2020)Brain tumor detection using fusion of hand crafted and deep learn-CognSystRes59:221–230

18. Sharif M, Amin J, Nisar MW, Anjum MA, Muhammad N, ShadSA (2020) A unified patch based method for brain tumor detectionCognSystRes59:273–286
19. Sharif M, Tanvir U, Munir EU, Khan MA, Yasmin M (2018)Braintumorsegmentationandclassificationbyimprovedbino-mialthresholding and multi-features selection. J Ambient IntellHumanComput:1–20.

20. Sharif MI, Li JP, Khan MA, Saleem MA (2020) Active deepneural network features selection for segmentation and recog-nition of brain tumors using MRI images. Pattern RecognLett129:181–189

21. SharifMI,LiJP,NazJ,RashidI(2020)Acomprehensivereviewonmulti-organstumordetectionbasedonmachinelearning.Pat-ternRecognLett131:30–37

22. Ohgaki H, Kleihues P (2013) The definition of primary and sec-ondaryglioblastoma.ClinCancerRes19:764–772

23. Cachia D, Kamiya-Matsuoka C, Mandel JJ, Olar A, CykowskiMD, Armstrong TS et al (2015) Primary and secondary gliosarco-mas: clinical, molecular and survival characteristics. J Neurooncol125:401–410

24. Amin J, Sharif M, Gul N, Yasmin M, Shad SA (2020) Brain tumorclassification based on DWT fusion of MRI sequences using con-PatternRecognLett129:115–122

25. SharifM,AminJ,RazaM,YasminM,SatapathySC(2020)An integrated design of particle swarm optimization (PSO) withfusion of features for detection of brain tumor. Pattern RecognLett129:150–157

26. Amin J, Sharif M, Anjum MA, Raza M, Bukhari SAC (2020) Con-volutional neural network with batch normalization for glioma andCognSystRes59:304–311

27. Amin J, Sharif M, Raza M, Saba T, Anjum MA (2019) Braintumor detection using statistical and machine learning method.ComputMethodsProgrBiomed177:69–79

28. Amin J, Sharif M, Gul N, Raza M, Anjum MA, Nisar MW et al(2020) Brain tumor detection by using stacked autoencoders indeeplearning.JMedSyst44:32

29. Johnson DR, Guerin JB, Giannini C, Morris JM, EckelLJ,Kaufmann TJ (2017) 2016 updates to the WHO brain tumorclassificationsystem:whattheradiologistneedstoknow.Radio-graphics37:2164–2180

30. WrightE,AmankwahEK,WinesettSP,TuiteGF,JalloG,CareyC et al (2019) Incidentally found brain tumors in the pediatricpopulation: a case series and proposed treatment algorithm. J Neu-rooncol141:355–361

31. PellegrinoMP,MoreiraF,ConfortoAB(2021)Ischemicstroke.Neurocritical care for neurosurgeons. Springer, New York, pp517–534

32. GarrickR,RotundoE,ChughSS,BrevikTA(2021)AcutekidneyEmergencygeneralsurgeryingeriatrics.Springer,NewYork,pp205–227

33. Lehmann ALCF, Alfieri DF, de Araújo MCM, TrevisaniER,NagaoMR,PesenteFS,GelinskiJR,deFreitasLB,FlauzinoT, Lehmann MF, Lozovoy MAB (2021) Carotid intima mediathickness measurements coupled with stroke severity strongly pre-dictshort-termoutcomeinpatientswithacuteischemicstroke:amachinelearningstudy.MetabBrainDis36:1747–1761

34. Scott AM (2005) PET imaging in oncology. In: Bailey DL,Townsend DW, Valk PE, Maisey MN (eds) Positron emissiontomography.Springer,London,pp311–325

35. WongTZ,vanderWesthuizenGJ,ColemanRE(2002)Positronemission tomography imaging of brain tumors. NeuroimagingClin12:615–626

36. WongKP,FengD,MeikleSR,FulhamMJ(2002)SegmentationofIEEETransNuclearSci49:200–207

37. BrennerDJ,HallEJ(2007)Computedtomography—anincreas-ingsourceofradiationexposure.NEnglJMed357:2277–2284

38. Smith-Bindman R, Lipson J, Marcus R, Kim K-P, Mahesh M,Gould R et al (2009) Radiation dose associated with commoncomputed tomography examinations and the associated lifetimeattributableriskofcancer.ArchInternMed169:2078–2086

39. Sinha, A., &Barde, S. (2022). Face recognition across age progression by using PCA. International Journal of Food and Nutritional Sciences, 11(S3), 4608-4617. https://doi.org/10.48047/IJFANS/S3/144

40. Sinha, A., &Barde, S. (2022). Age Invariant Face Recogntion Using Pca And Msvm. Journal of Pharmaceutical Negative Results, 2174-2185.

41. Sinha, A., &Barde, S. (2022, October). Illumination invariant face recognition using MSVM. In AIP Conference Proceedings (Vol. 2455, No. 1, p. 040005). AIP Publishing LLC.

42. Pandey, Akash& Sinha, Anupa. (2023). Journal of Clinical Otorhinolaryngology, Head, and Neck Surgery BRAIN TUMOR DETECTION USING CANNY EDGE DETECTOR WITH MACHINE LEARNING. 2023.