(PDF) A Review of Applications of Artificial Intelligence in Heavy Duty Trucks

How AI Can Predict, Diagnose And Track Infectious Outbreaks In Real Time

A new review published in Viruses explores the transformative potential of machine learning (ML) and artificial intelligence (AI) in the surveillance, diagnosis, and prognosis of infectious diseases. Titled "Machine Learning and Artificial Intelligence for Infectious Disease Surveillance, Diagnosis, and Prognosis", the paper delivers a comprehensive analysis of how these technologies can reshape public health systems in the face of ongoing and future pandemics.

The research systematically dissects the role of ML and AI across three key domains: early detection and surveillance of emerging infectious diseases, clinical diagnostics and decision support, and prognostic modeling to predict disease outcomes. This integrative approach sets a new benchmark in understanding how data-driven frameworks can be deployed to improve epidemic response strategies globally.

How can machine learning enhance disease surveillance and outbreak detection?

The study begins by establishing the crucial role of ML in public health surveillance. It identifies that early warning systems for infectious diseases often rely on fragmented datasets, ranging from climate and mobility data to socio-economic and demographic inputs, that traditional epidemiological models struggle to integrate. ML offers a solution by processing these multi-modal datasets with high speed and precision.

The authors emphasize that ML-driven surveillance systems can flag early signals of disease outbreaks by detecting non-linear trends and hidden correlations in data streams. For instance, models trained on environmental conditions like temperature and rainfall have demonstrated efficacy in predicting vector-borne diseases, such as dengue. These applications could be particularly valuable in low- and middle-income countries (LMICs), where infrastructure for early detection is limited.

Furthermore, real-time mobility data and digital exhaust from internet activity, such as search engine queries and social media posts, are becoming integral to outbreak forecasting. ML algorithms can sift through these dynamic datasets to track changes in population behavior, symptom reporting, and disease sentiment, offering authorities a lead time to implement containment measures.

However, the authors caution that the success of such systems is contingent on high-quality input data. Issues like incomplete datasets, privacy concerns, and lack of standardization can undermine the reliability of predictions. Still, the study finds compelling evidence that ML-powered surveillance can outperform conventional models, especially in scenarios with rapidly evolving pathogens like SARS-CoV-2.

What role does AI play in diagnosis and risk stratification?

Moving from detection to diagnosis, the study explores how ML algorithms can optimize the clinical workflow. ML models can integrate complex diagnostic inputs, including clinical history, blood markers, imaging results, and symptoms, to assist healthcare providers in triaging and diagnosing patients more accurately and efficiently.

The authors highlight that ensemble learning methods, particularly those using deep learning architectures such as convolutional neural networks (CNNs), random forests, and support vector machines (SVMs), have shown promising results in identifying disease severity and stratifying patient risk. These models not only reduce the time required for diagnosis but also minimize human error, especially under the pressure of high caseloads during pandemics.

Importantly, the paper advocates for the use of Explainable AI (XAI) to ensure transparency in clinical decision support tools. By offering visual or numerical explanations for predictions, XAI helps clinicians validate and trust the output of ML systems. This is essential for real-world application, where accountability and interpretability are critical.

The research also emphasizes the need for AI tools to complement, not replace, human decision-making. While algorithms can provide data-driven insights, final diagnostic decisions must rest with trained clinicians who can interpret predictions within the broader clinical context.

How are prognostic models shaping the future of personalized medicine?

The study also focuses on prognosis: forecasting disease progression and outcomes in infected individuals. ML models trained on historical and real-time patient data can identify biomarkers and clinical patterns associated with severe disease, helping clinicians anticipate complications and personalize treatment plans.

This feedback mechanism between diagnosis and prognosis is framed as a continuous loop facilitated by reinforcement learning, wherein the outcomes of previous cases are used to refine future predictions. The study underscores that such iterative learning is pivotal for the evolution of precision medicine, particularly for diseases with variable clinical presentations.

The authors point to the integration of patient-specific features like age, co-morbidities, and immune response markers into ML models, which enhances the precision of prognostic predictions. Tools like SHAP (Shapley Additive exPlanations) and Grad-CAM (Gradient-weighted Class Activation Mapping) are noted for their contribution in interpreting model outputs at a granular level, making them valuable in tailoring patient care.

Four Ways AI And Machine Learning Will Transform Healthcare In 2020

From AI assisted robotic surgery to clinical diagnosis, image analysis and administrative tasks, the use of artificial intelligence and machine learning (ML) technologies is increasing within healthcare. In fact, 75% of healthcare enterprises are planning to execute an AI strategy next year, whether that's exploring how it can automate critical but repetitive tasks to free up time for clinicians, how automatic speech recognition can speed up disease diagnosis, or how it can create synthetic controls for clinical trials. 2020 holds great opportunity to further unleash its potential.

With this in mind, we spoke to several industry experts operating in the AI and ML space for healthcare and life sciences, to explore the key trends and challenges we can expect to encounter in 2020.

More effective deployment of AI tools will become a focus

Firstly, when executing an AI strategy, organisations within the industry will start to focus to a greater extent on where AI/ML technologies can be used most effectively, and how to deploy them for maximum benefit in real clinical scenarios.

This will require careful consideration around how tools are implemented across various areas of healthcare, and for Mario Nacinovich, global head, communications & marketing at AiCure – which uses AI to see, hear and understand how patients respond to treatment – there are different challenges associated with each area.

"From a societal standpoint, building greater trust in AI and protecting personal healthcare data will continue to be among the omnipresent challenges. From an administrative standpoint, making it easier for AI to integrate with existing technology infrastructure will certainly help adoption."

"AI will play a critical role in understanding how a drug is performing in real-time and how patients are responding in clinical research including medication adherence and their behaviour"Mario Nacinovich

Overarchingly, though, he believes that deploying AI capabilities more effectively comes down to "ensuring that back-end processes gain greater efficiencies" in order to reduce timescales and provide better outcomes for patients.

The idea of effective deployment is also one which Eyal Gura, CEO and co-founder of Zebra Medical Vision, an imaging analytics platform which uses AI, believes will become a focus in 2020: "Having a single AI solution that integrates seamlessly into existing workflows at an affordable rate will be critical in supporting clinicians in delivering better patient care."

As Gura explains, with two billion people joining the middle class, a rising ageing population and the growing shortage in medical experts, "AI will be critical in enabling communities to provide productive and consistent health services".

AI will be increasingly applied to imaging diagnostics

Charles Taylor, co-founder of HeartFlow, which has developed a noninvasive test that helps clinicians to understand the severity of coronary heart disease, believes that we are only just beginning to see the full benefits of what medical imaging and AI can do for diagnostics.

"Right now, we're able to use medical imaging and AI to give physicians unprecedented insight into potentially life-threatening restrictions on blood flow within the body," he says. "But we've only just scratched the surface of what integration between information technology, computers and healthcare can achieve, and the expectations are high."

Dr Michalis Papadakis, CEO and co-founder of Brainomix – spun out of the University of Oxford – also believes we can expect to see AI and ML "become the driving force behind imaging diagnostics".

"With around 780,000 people suffering a stroke each year in Europe, and 7.4 million people living with heart and circulatory diseases in the UK, it is imperative we find ways to reduce the burden on healthcare organisations and improve time to disease detection.

"The number of MRI and CT scans, for example, is already on the rise, and AI has the ability to read scans as accurately as an expert physician. Utilising these new technologies to review scans for any disease can reduce patient wait time and ease the burden on medical staff. There will be greater recognition next year of the value of AI in augmenting human performance."

Novel and digital biomarkers will support disease diagnosis

Being able to unlock new biomarkers is crucial in diagnosing diseases, and AI/ML technology has the power to enhance the use of biomarkers. One area that biomarkers are being used to great effect is in the diagnosis of Alzheimer's and Dementia. Dr Steven Chance, CEO at Oxford Brain Diagnostics, explained that the company is focused on improving the treatment of Alzheimer's by using ML to analyse MRI scans of the cerebral cortex, enabling early and accurate detection. He says: "Dementia remains highly complex in nature and requires extensive collaboration to succeed. Urgent action to address these challenges is needed today."

He also explains that "unlocking new biomarkers, leveraging smarter science and deploying funds where they are needed most may give the industry a chance to defeat the terrible condition. We must re-focus our efforts and move quickly now towards examining the disease much earlier, allowing novel biomarkers to measure the progression more accurately and develop specific and targeted drug treatments for the range of dementias that exist."

Francesca Cormack PhD, director of research & innovation at Cambridge Cognition, is also focused on digital biomarkers and the use of AI/ML for detecting neurodegenerative conditions such as Alzheimer's.

"The upward trajectory of digital capabilities over the last decade, combined with the widespread adoption of devices, has augmented biological markers with digital measures of disease progression," she says.

"In our field, it is now possible to use AI to enrich cognitive test scores with metrics that indicate cognitive effort, i.E. The unique features of a patient's voice that reveal when they are finding it particularly challenging to perform a task. Patients who are ostensibly performing within normal ranges but struggling to maintain that performance are likely suffering with the early stages of decline and could benefit from interventions that might slow or prevent further neurodegeneration.

"Over the next year, we expect to see improvements in the precision of digital biomarkers for rapidly detecting neurodegenerative conditions such as Alzheimer's disease. The ultimate goal is to integrate digital biomarkers into clinical care and improve patient outcomes."

Clinical trials will continue to be transformed

Finally, there is huge potential for AI and ML to continue to transform clinical trials. As Mario Nacinovich explains: "Once identified and recruited, one of the biggest challenges in clinical trials is keeping subjects engaged and optimised to treatment. Medication non-adherence has been shown to increase variance, lower study power, and reduce the magnitude of treatment effects. AI will play a critical role in understanding how a drug is performing in real-time and how patients are responding in clinical research including medication adherence and their behaviour.

"The adoption of new technologies in 2020 and beyond have the potential to provide clinicians with improvements in overall patient engagement, outcomes, quality of life, practicality in use, and reduce clinical development time and associated costs."

To conclude, Eyal Gura's insights provide a concise summary of what we can expect to see from the AI and healthcare industries in the near future: "From medical imaging analysis to sensors and smart alerts, we are going to witness more improved and personalised care. In 2020, we will see AI in deployment of hundreds of health networks globally and impact on millions of patient lives. AI has the power to transform patient care and empower radiologists to help with patient diagnosis."

New Machine Learning Program Accurately Predicts Who Will Stick With Their Exercise Program

Researchers used machine learning to predict who sticks with exercise by analyzing lifestyle and body data from over 11,000 people. (CREDIT: CC BY-SA 4.0)

Staying active is one of the most important things you can do for your health. Regular exercise helps you live longer, lowers your risk of disease, improves your mood, and boosts energy levels. But only a small portion of people actually meet exercise recommendations. So, what makes someone stay committed to working out?

A team of researchers set out to find the answer. At the University of Mississippi, scientists analyzed national health data using machine learning to find patterns in who meets physical activity (PA) guidelines and why. This approach could help doctors and trainers better support your health by understanding what motivates people like you to keep moving.

This study, published in the journal, Scientific Reports, looked at data collected between 2009 and 2018 from the National Health and Nutrition Examination Survey, a large U.S. Survey that tracks health and diet habits. The research team included doctoral students Seungbak Lee and Ju-Pil Choe, and Professor Minsoo Kang. They used a tool called machine learning to sort through over 30,000 survey responses.

Machine learning helps computers find patterns in large amounts of data. Unlike older statistical tools, which expect clean, linear data, machine learning works well even when the data is messy or complicated. It can sort out which pieces of information matter most in predicting behavior, like who sticks with exercise routines.

Creation of new variables. Vigorous work and recreational activity minutes were doubled because the established PA guidelines consider 75 min of vigorous-intensity activity equivalent to 150 min of moderate-intensity activity. (CREDIT: Scientific Reports)

The researchers filtered the data to include only people age 18 and older without diseases that could limit exercise, such as cancer, diabetes, or arthritis. After removing entries with missing answers, the final data set included 11,638 participants.

Each person's responses were grouped into three main areas: demographics (age, gender, race, income, etc.), body measurements (like body mass index and waist size), and lifestyle habits (such as alcohol use, smoking, sleep, and sedentary time). The goal was to build models that could predict whether someone met the weekly activity guidelines.

According to U.S. Health officials, adults should get at least 150 minutes of moderate exercise or 75 minutes of intense activity each week. Unfortunately, the average American only gets about two hours of activity weekly—half of what's recommended.

Using six different machine learning algorithms, the researchers built 18 prediction models to test various combinations of factors. These models were measured by how accurate they were, how well they could find patterns, and how balanced their predictions were.

Related Stories

The best-performing model was a decision tree using all available variables. It had an accuracy of about 70.5% and an F1 score (a balance between precision and recall) of 0.819. That means it correctly predicted who met exercise guidelines most of the time.

But beyond just performance, the team wanted to know which specific factors were most useful in making predictions. Using a technique called Permutation Feature Importance (PFI), they found that sedentary behavior, age, gender, and educational status were the most important predictors. Even though some models gave slightly different answers, these factors kept showing up again and again.

Ju-Pil Choe explained, "I expected that factors like gender, BMI, race or age would be important for our prediction model, but I was surprised by how significant educational status was. While factors like gender, BMI and age are more innate to the body, educational status is an external factor."

The team noted that people who sat for long periods, had lower education levels, or were of a certain gender were less likely to meet activity guidelines. This helps explain who is more likely to stick with physical activity and why. These insights could guide future programs aimed at helping people develop healthier habits.

Confusion matrix. (CREDIT: Scientific Reports)

While the results are promising, the researchers did note some limits to their approach. One key issue is that the survey data relied on self-reported activity levels. People often overestimate how much they exercise when asked to recall it from memory.

"One limitation of our study was using subjectively measured physical activity data," Choe said. "More accurate, objective data would improve the study's reliability."

Future research could fix this by using wearable fitness trackers or apps that automatically log physical activity. Machine learning could then use that objective data to find even stronger and more detailed patterns.

Despite this limitation, the research shows that machine learning has great promise for studying health behaviors. It doesn't just tell us what the trends are—it helps uncover why those trends exist in the first place.

Why is all this important? Because understanding the reasons behind someone's exercise habits can help health professionals create better, more personalized plans. Instead of giving the same advice to everyone, doctors could use data-driven models to figure out what motivates each person.

Top 5 features in top 10 models. Different color density indicates weight difference (high density means higher importance). BMI body mass index, ES educational status, SB sedentary behavior, PFI permutation feature importance, WC waist circumference. (CREDIT: Scientific Reports)

For example, if someone has a sedentary job and low education levels, they might need more support or different types of motivation to stay active. Knowing that these factors matter allows experts to build programs that work for each individual.

This is especially helpful for trainers, coaches, and even health app developers. They can create exercise routines that feel more achievable and are tailored to your lifestyle, age, and daily habits. It makes sticking with a workout plan easier and more realistic.

Professor Kang summed up the purpose of the study: "Physical activity adherence to the guidelines is a public health concern because of its relationship to disease prevention and overall health patterns. We wanted to use advanced data analytic techniques, like machine learning, to predict this behavior."

Other studies have already used machine learning in related ways. For example, some researchers built models to classify physical activity in children using motion sensors. Others used neural networks to sort activity levels based on body movements. But this study is one of the first to focus on predicting adherence to activity guidelines using only self-reported data and a wide mix of demographic, body, and lifestyle factors.

The results show that machine learning can be a powerful tool in public health. It reveals patterns that might be invisible with traditional methods. And it gives researchers a new way to help people live healthier, longer lives.

Note: The article above provided above by The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.

---