At the 35th AICC-RCOG Annual Conference in association with FOGSI and MOGS, Dr. Niranjan Chavan, President of MOGS, gave an address on Artificial Intelligence in Gynaecologic Oncology at Taj Lands' End, Bandra, Mumbai on the 6th November 2022
High Profile Call Girls Coimbatore Saanviâď¸ 8250192130 Independent Escort Se...
Â
AI in Gynaec Onco
1. AI IN GYNAEC ONCOLOGY
35th AICC RCOG Annual Conference,
Taj Lands End, Mumbai
6th November 2022
2. Professor and Unit Chief, L.T.M.M.C & L.T.M.G.H, Sion Hospital
President, MOGS (2022-2023)
Joint Treasurer, FOGSI (2021-2024)
Member Oncology Committee, SAFOG (2020-2021) (2021-2023)
Dean AGOG & Chief Content Director, HIGHGRAD & FEMAS Courses
Editor-in-Chief, FEMAS, JGOG & TOA Journal
64 publications in International and National Journals with 120 Citations
National Coordinator, FOGSI Medical Disorders in Pregnancy Committee (2019-2022)
Chair & Convener, FOGSI Cell Violence Against Doctors (2015-16)
Member, Oncology Committee AOFOG (2013-2015)
Coordinator of 11 batches of MUHS recognized Certificate Course of B.I.M.I.E at L.T.M.G.H
(2010-16)
Member, Managing Committee IAGE (2013-17), (2018-20)(2022)
Editorial Board, European Journal of Gynaec. Oncology (Italy)
Course Coordinator of 3 batches of Advanced Minimal Access Gynaec Surgery (AMAS) at LTMGH
(2018-19)
DR. NIRANJAN CHAVAN
MD, FCPS, DGO, MICOG, DICOG, FICOG, DFP,
DIPLOMA IN ENDOSCOPY (USA)
3.
4.
5. WHAT IS AI?
⢠Artificial intelligence (AI) is a type of digital computer
system that parallels the way the human brain processes
information.
⢠AI is organized in a similar way that neurons in the brain are
arranged, with their multiple neural nodes, and so are
referred to as neural networks.
⢠The rise of AI has led to the subsequent development of
artificial neural networks (ANN), which consist of a
dependable mathematical system that can interpret
multifactorial data.
6. ⢠These neurons are connected via multiple synapses and
send the data to each other back and forth, and by doing
so, come up with the most probable answer.
⢠Making these multiple connections enables computers to
mimic cognitive functions, such as the reasoning
process, to identify the most probable answer to a
problem.
7. ⢠This complex algorithm AI software is now utilized in medicine to analyze large
amounts of data, which can assist in disease prevention, diagnosing, and
monitoring patients.
⢠Overall, AI can aid practitioners in decision-making and will help clinicians to
make more self-assured decisions.
8.
9. WHAT IS MACHINE LEARNING ?
⢠ML, is a form of AI, in which a machine can
learn and adapt to situations and undergo self-
driven data training.
⢠Typically, a training data set is used to train a
computer program by feeding images
describing a series of features such as colour,
shape, and texture.
⢠Two main approaches to ML, viz supervised and
unsupervised learning.
10. ARTIFICIAL NEURAL NETWORKS
⢠A neural network typically consists of several layers of
artificial neurons, fully connected to each other.
⢠Each neuron receives signals from multiple neurons from
the previous layer, integrates these signals, and then
fires these integrated signals, in all directions.
⢠ANNs are mathematical systems which are reliable,
flexible and evaluate multifactorial data at lightening
speed.
11.
12. AI IN OBGYN
⢠Fetal Heart Rate Monitoring and Pregnancy Surveillance
⢠GDM
⢠Preterm Labour & AI in Ultrasound
⢠IVF
⢠Urogynecology
⢠Gynaec Oncology
⢠Parturition
14. AI IN OVARIAN CANCER
⢠In this study, a new technology for automatic identification of adnexal masses based on a neural network (NN)
method was tested
⢠They calculated seven different types of features (local binary pattern, entropy, law texture energy, invariant
motion, gray level co-occurrence matrix, Gabor wavelet and fractal dimension) from ultrasound images of the
ovary, extracted several parameters from these features and collected them together with the age of
gynecological patients
Ultrasound in Medicine and Biology; 2016;42 (3):742â752
15. ⢠It suggests that the machine learning system, especially integrated classifiers, can provide critical diagnosis
and prognosis prediction for patients with EOC before initial intervention, and the use of predictive algorithm
can promote personalized treatment selection and avoid âone-size fits allâ treatment approach through pre-
treatment stratification of patients.
⢠The study also pointed out that in order to further improve the accuracy of prediction, AI should be used in
future studies to determine the prediction characteristics of preoperative blood value time series.
April 2019
16. ⢠By using this Computer Aided Design technology, which based on a Neural
Network, they found the accuracy of automatic identification of malignant adnexal
masses was 98.78%, the sensitivity was 98.50%, the specificity was 98.90%, and
the area under the receiver operating characteristic curve (AUC) reached 0.997
⢠The highlight of this study lies in considering a wide range of texture features
and making use of advantages that NNs can generalize
17. ⢠No screening for ovarian cancer exists despite it
being a common gynaecological cancer.
⢠Thus, most cases are diagnosed in advanced
stages, leading to a high five-year mortality
rate.
⢠Researchers at Brigham and Womenâs Hospital
and Dana-Farber Cancer Institute have been
using AI to manipulate large amounts of micro
ribonucleic acid (RNA) data to develop models
that can potentially diagnose early ovarian
cancer.
18. ⢠The AI neural network was able to keep up
with the complex interactions between micro
RNA and accurately identified almost 100%
of abnormalities that represented ovarian
cancer,
⢠as opposed to an ultrasound screening test
that was able to identify abnormal results less
than 5% of the time.
19. AI IN ENDOMETRIAL CANCER
⢠Three different methods including logistic regression, artificial neural networks (ANNs)
and Classification and Regression Tree(CARTs) were used to compare diagnostic accuracy
of endometrial carcinoma in postmenopausal women presenting with endometrial
thickness of 5 mm or abnormal vaginal bleeding
Public Health. 2018;164:1â6
20. ⢠The diagnostic accuracy of three methods was determined by ultrasonography combined
with the final pathological results. The study found that the sensitivity of the ANN model
(86.79%) was much higher than that of CARTs (78.3%) and logistic regression model
(p<0.05 in both comparisons)
⢠AI, especially Deep Learning with particularly high sensitivity and specificity, is a powerful
and useful mathematical tool, which can be used in primary health care and considerably
promote public health
21. ⢠Texture analysis was performed using commercial research software and the surrounding region of
interest was manually delineated. The results show that texture analysis and radio-frequency
modelling based on MRI can accurately diagnose the presence of
⢠Deep Myometrial Invasion,
⢠LVSI, and
⢠high-level tumours
Radiology. 2017;284(3):748â757
22. ⢠Thus, texture features based on MRI can be used for computer-aided diagnosis,
and MRI combined with AI can distinguish clinical pathologic prognosticators
before treatment thus providing enough clinical benefit for patients with
endometrial cancer
⢠These preliminary results indicate that the proposed way to obtain a promising
discriminative power can be used together with conventional MRI sequences to
distinguish sarcomas from myomas.
23. ⢠For noncancerous and cancerous cervical images, the proposed system achieved classification
accuracy of 97.14% and 100%, respectively
⢠This proposed methodology for cervical image classification achieved 98.57% of the total
classification accuracy
Asian Pac J Cancer Prev. 2018;19(11):3203â3209
AI IN CERVICAL CANCER
24. ⢠The performance analysis of the proposed cervical cancer detection and
segmentation system showed that its sensitivity was 97.42%, specificity was
99.36%, and segmentation accuracy was 99.36%.
⢠Therefore, the simulation on these cervical image data sets shows that the new
method is superior to the traditional cervical cancer detection and segmentation
methods and has higher performance in clinical practice
25. ⢠This technology can automatically extract image patches coarsely centred on the nucleus as
network input, which means that it can extract deep features embedded in cell image blocks for
classification
⢠It was found that this method yielded the highest performance not only on the Herlev Pap
smear, but also on the H&E staining manual liquid-based cytology (HEMLBC) liquid-based
cytology datasets
⢠Therefore, it is expected that this type of cervical cell classification system with segmentation-
free and high accuracy will be developed into an automatic assisted reading system for primary
cervical screening
26. Computer Methods and Programs in Biomedicine 138 (2017) 31â47
⢠Classified Pap smear images in their research by using the integrated classifier which
was designed with three popular individual classifiers: SVM, neural network
multilayer perceptron (MLP) and random forest classifier (RF)
⢠All features are proved that it is very important for the classification of Pap smear
samples, and a single feature cannot provide high accuracy in the classification
process
⢠The study also found that the performance of the ensemble classifier is the best, and
the performance of MLP and SVM are similar, both of which are better than RF
27. ⢠The current screening consists of visual
inspection of the specimen collected during a
Papanicolaou (PAP) smear and using acetic acid
to visualize whitening in the tissue which would
be indicative of disease.
⢠Despite its convenience and low cost, it lacks
accuracy.
⢠AI has outperformed human experts in
interpreting cervical pre-cancer images.
28. AI IN DRUG RESEARCH AND
DEVELOPMENT
⢠With the individual differences of malignant tumour patients and the emergence of
multidrug resistance, many patients with gynaecological cancer have poor drug sensitivity
resulting in unsatisfactory clinical treatment results
⢠A pharmacodynamic model of the anticancer potential of synthetic compounds was
established, and it was found that the therapeutic effect could be optimized by adjusting the
drug efficacy and response heterogeneity through changing the exposure time
⢠It is believed that this method may effectively infer the in vitro results of lead compounds
produced both in vivo and preclinical research
Computational Biology and Chemistry Volume 79, April 2019, Pages 137-146
29. ⢠Watson for Oncology (WFO), an AI computer program, was developed by IBM
Corporation (USA) with the help of top oncologists from Memorial Sloan Kettering
Cancer Center (MSK)
⢠Their goal was to create a cognitive computing system to meet todayâs big data
information challenges
⢠The scientists who developed WFO integrated natural language processing,
information retrieval, knowledge expression, machine learning, and general
reasoning modes, acquired and evaluated a large amount of structured and
unstructured data from previous medical records through machine learning and
natural language processing to make recommendations for cancer treatment
AI IN CLINICAL DECISIONS
30. ⢠As for supported cases, the treatment recommendations
provided by WFO are divided into three groups:
⢠green buckets - recommended, which represents a treatment
supported by obvious evidence;
⢠yellow buckets - for consideration, which represents a
potentially suitable alternative; and
⢠red buckets - not recommended which stands for a treatment
with contraindications or obvious evidence against its use.
⢠Although WFO helps to reduce the time required for clinical
treatment planning, due to the recent development of cognitive
computing technology, there is still a lack of large-scale data
applied outside the US
31. ⢠Advantages -
⢠it improves doctorsâ work efficiency and reduces workload
⢠it can prevent man-made calculation errors. Chemotherapy scheme and drug selection
involve multiple clinical formulas, which need to be calculated one by one in sequence.
⢠it can improve the quality of doctorâpatient communication and prevent doctorâpatient
disputes
⢠Limitations -These are mainly related to
⢠different drug choices,
⢠different treatment options,
⢠coexisting diseases of patients,
⢠economic factors in different countries
32. AI IN PROGNOSIS
⢠Neural network models are being used to deliver
prognoses in patients with ovarian cancer.
⢠In a report done by Enshaei et al. 2015, ANN was able
to predict survival with a 97% accuracy.
⢠The AI systems they developed have the potential of
providing an accurate prognosis.
33.
34. ⢠Norwitz et al in 2015 have created an AI software that can
predict prognosis in patients with ovarian cancer more
precisely than current method.
⢠It can also predict the most effective treatment according
to the diagnosis of each patient.
⢠Long-term survival rates for advanced ovarian cancer are
poor; thus, more targeted therapies are needed.
35. TAKE HOME MESSAGE
⢠AI has a promising future in overcoming diagnostic challenges and improving treatment
modalities and patient outcomes in OBGYN
⢠Further studies need to be done to decrease bias when creating algorithms and to increase
adaptability in the system, enabling the incorporation of new medical knowledge as new
technology surfaces
⢠AI is not meant to replace practitioners but rather to serve as an adjunct in decision-making
⢠Clinicians must embrace them, yet be wary, and when necessary, recognize its advantages
and drawbacks to continue providing the best patient care
⢠AI is properly used and its applications in clinical practice are optimized,67 it will be regarded
as a valuable tool
⢠AI cannot only be used as a promising tool in gynecologic malignant tumors, but also as a
method to resolve several long-term challenges.