Artificial intelligence (AI) in medicine is advancing, giving doctors new tools to offer better patient care recommendations, diagnostic assistance, treatment suggestions, and the latest medical knowledge. AI, which consists of models and algorithms capable of summarizing existing data to produce future predictions, has the potential to revolutionize many aspects of our lives. AI-driven methods could potentially improve the prognosis and treatment of Alzheimer’s disease (AD). Improving early identification of AD can improve patient outcomes and quality of life by using machine learning techniques and incorporating chat robots (Chatbots). Chatbots, one of the many applications of AI, are virtual conversational agents that allow users to engage with AI-based computer programs through voice or writing. Recently, chatbots have been launched in various fields, including business, retail, and services such as healthcare. Existing research has demonstrated the effectiveness of medical chatbots in activities such as ensuring proper medication adherence. Cancer patient satisfaction has been shown to increase after having a conversation with a chatbot. Interacting with an empathetic chatbot has a mitigating impact on patients’ mental health. Interactions with medical chatbots reduced symptoms of hopelessness and anxiety, while medical chatbots correctly identified more than 96% of infected patients during the COVID-19 pandemic [11]. The burden on healthcare practitioners can be reduced and simplified with the help of medical chatbots, such as AD predictive chatbots.
AD is a degenerative brain disease that causes memory loss and cognitive impairment, including difficulty speaking, thinking, and completing tasks. AD is named after Alois Alzheimer, who first discovered it in 1906. AD causes 60-80% of all dementia cases. In 2020, approximately 57.4 million people were diagnosed with dementia. In other words, 2 out of 3 people with dementia have AD. The increase in the number of AD sufferers is expected to reach 152 million by 2050. Individuals affected by AD experience a severe decline in cognitive function, and this has a significant impact on quality of life and general health. In addition, the average age of Alzheimer’s sufferers after diagnosis is 7.6 years and 5.8 years. Mild cognitive impairment (MCI) is a promising stage because it is still in the preclinical stages of AD, serving as a specific target for early treatment with the potential to halt or slow AD progression.
This suggests that MCI is an effective early-stage intervention to reverse or halt the pathological progression of AD. Magnetic resonance imaging (MRI) can provide comprehensive 3D images of internal body components such as the brain. MRI has been widely used to understand morphological and functional brain changes in vivo, including AD, schizophrenia, and others. Therefore, structured MRI can provide information about the anatomical structure of the brain, helping in detecting and quantifying AD brain shrinkage patterns. This research expands previous research that used brain MRI imaging and machine learning to predict the onset of AD. To improve prediction accuracy, this research combines convolutional neural network (CNN) and support vector machine (SVM) algorithms in a new way in a chatbot platform. This research also looks at how preprocessing methods affect model performance. Overall, this research advances the field of AI in healthcare by investigating cutting-edge methods to improve patient care and early diagnosis of AD.
This research explores nine combination schemes between 3 preprocessing methods and 3 proportion datasets in the CNN-SVM model. The three preprocessing methods are 1 without preprocessing and 2 preprocessing. The evaluation results show that the best performance is obtained in the model with scheme four based on accuracy, precision, recall, and F1-score, each reaching values of 98%, 99%, 98%, and 98%, respectively. In other words, the best pre-processing mechanism among the three studied schemes involves resizing the image to a size of 150 pixels×150 pixels, followed by scaling the image to the range 0-1. The choice of this preprocessing scheme shows that the specific steps implemented in preprocessing one make a positive contribution to the model performance. Resizing images to smaller dimensions and scaling to change pixel values to more concentrated ranges have improved the model’s ability to extract features and understand brain MRI image patterns. On the other hand, the chatbot has demonstrated excellent functionality in providing AD-related responses after transition testing, with a functionality score reaching 99.64 points out of 100.
Author: Suryani Dyah Astuti and Dezy Zahrotul Istiqomah Nurdin
Detailed information from this research can be seen in our article at:
Here’s my take on the article, with a dash of humor and a pinch of sarcasm:
Artificial Intelligence in Medicine: Because Doctors Needed More Robots Telling Them What to Do
You know, folks, AI is getting so good at everything, I’m starting to think we’ll soon have robots doing our laundry, making our coffee, and even writing our jokes. I mean, who needs humans when you have algorithms and machine learning? (Just kidding, I’m not worried about my job… yet.)
Seriously though, AI in medicine is a game-changer. It’s like having a super-smart, never-tiring medical student who can analyze data, make predictions, and even chat with patients (just don’t expect it to make a decent cup of tea). This article highlights some exciting advancements in AI-driven methods to improve early detection and treatment of Alzheimer’s disease.
Alzheimer’s: The Memory-Thieving Bandit
Alzheimer’s is a nasty piece of work, folks. It’s like a sneaky thief that steals your memories and swaps them with… well, nothing. And before you know it, you’re wondering where you put your keys (or your kids). According to the stats, 60-80% of all dementia cases are caused by Alzheimer’s, affecting a whopping 57.4 million people in 2020. That’s a lot of people who might be wondering what day it is (spoiler alert: it’s probably Tuesday).
Enter the Chatbots: The Siri of Healthcare
Now, I know what you’re thinking: "Chatbots? Aren’t those just annoying little boxes that pretend to care about my shopping experience?" Well, yes and no. In healthcare, chatbots are being used to actually help patients, not just sell them stuff. These AI-powered conversational agents can assist with medication adherence, identify symptoms, and even offer empathy (wow, robots with feelings – who knew?). One study even showed that interacting with a medical chatbot reduced symptoms of hopelessness and anxiety. Who needs a shrink when you have a robot?
The Science-y Bit
Okay, let’s get to the meat of the article. Researchers Suryani Dyah Astuti and Dezy Zahrotul Istiqomah Nurdin have been working on a neat project that combines convolutional neural networks (CNNs) and support vector machines (SVMs) to improve the accuracy of Alzheimer’s diagnosis using brain MRI images. They found that the best performance was achieved by resizing images to a certain size (150 pixels x 150 pixels – don’t ask me why, I’m not a scientist) and scaling the image to a range of 0-1. It’s all very technical, but essentially, they made a robot that’s really good at looking at pictures of brains and spotting the bad guys (the Alzheimer’s-causing kind).
The Verdict
All in all, this research is a significant step forward in using AI to improve patient care and early diagnosis of Alzheimer’s disease. And if you’re thinking, "This is all well and good, but what about the humans?", fear not! These chatbots are designed to help doctors, not replace them. Yet.
So, there you have it – AI in medicine: because who needs human intuition when you have algorithms and machine learning? (Just kidding, sort of.) Seriously, this research has the potential to make a real difference in people’s lives. Now, if you’ll excuse me, I need to go ask my AI assistant to write a joke for me… just kidding (or am I?).