fbpx

Interview with consultant histopathologist Jan von der Thüsen MD PhD

By News

“The solution offered by Aiforia for AI in pathology is extremely user-friendly and while it relies heavily on correct annotation by disease specialists, it assumes very little prior knowledge of AI – a perfect combination for me. “

What do you do and what is your research focus?

I am a thoracic pathologist and my research interests include both neoplastic and non-neoplastic diseases of the thorax, including lung cancer, mesothelioma, thymoma, interstitial lung disease and lung and heart transplantation.

Why did you apply to the aiForward program?

The solution offered by Aiforia for AI in pathology is extremely user-friendly and while it relies heavily on correct annotation by disease specialists, it assumes very little prior knowledge of AI – a perfect combination for me. Also, the online support offered by knowledgeable Aiforia staff is extremely helpful, including very useful instruction videos.

Had you thought much about using AI in your research before applying to aiForward?

No, while we had been considering entering the field for a while, participating in an AiForward program has really kick-started our AI research.

Tell us a little bit about your aiForward project.

Classification of rejection in solid organ transplantation is subject to interobserver variation. In an effort to reduce the variability and to improve correlation with outcome, as well as a point to potentially useful subclassifications of current categories, we have started an aiForward program to score kidney and heart biopsies according to current guidelines.

What are your expectations for the program?

To develop an algorithm which reproduces current transplant rejection scoring systems, as well as (hopefully) provides a more accurate correlation with clinical outcome.

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

Artificial intelligence for Parkinson’s disease

By News

AI can automate it and make it much easier and more reproducible.”

Ilmari Parkkinen is on a mission to decipher Parkinson’s disease (PD). The neurodegenerative disorder predominantly affects neurons that produce dopamine. Current treatments alleviate the symptoms caused by the loss of these neurons. Ilmari’s research is focused on elucidating why these particular neurons degenerate and ultimately to discover disease-modifying treatments to either halt or reverse the progression of the disease. 

Preclinical PD studies often involve numerous quantitative analyses, such as cell counting. This is mostly done manually with labor intensive methods like stereology. Hours and days are spent counting cells with this traditional method, often involving multiple scientists. This, like for many other neuroscientists, was a big hurdle for Ilmari and his lab. 

Neuron counting by AI

“AI can automate it and make it much easier and more reproducible,” explained Ilmari of why he decided to apply to the aiForward program. “Because Aiforia offered a chance to create a custom algorithm just for this purpose, we decided to go for it and use it in our models,” he added.”

Aiforia Create allows Ilmari to develop his own AI model not just to count a huge amount of cells but to also identify and quantify the specific cells he needs to analyze for his Parkinson’s disease study. A customized algorithm was needed and Aiforia Create enabled the development of this. “Using AI for your research requires a good amount of expertise and/or a good platform to get started,” he described.

Ilmari has finished developing his algorithm and is now excited to start running the analysis, he recently told us: “Mainly we are expecting to do exciting and robust work that could lay the basis for our future studies to study the mechanisms underlying neurodegeneration. Also, at the end of the day, hopefully we will find a way to make the lives of neuropathologists and researchers doing preclinical studies on PD easier. The great thing is that we already have some promising preliminary data.”

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

Interview with PhD student Polina Stepanova

By News

“The AiForward program can give the opportunity to do image analysis faster and more unbiased.”

What do you do and what is your research focus?

I am a PhD student at the Voutilainen Lab at the University of Helsinki’s Institute of Biotechnology. One of the interests of our lab is the identification of suitable drug treatment for Huntington´s disease (HD). HD is a fatal inherited neurodegenerative disease. I am focusing on the research of the effect of a promising protein on HD at the preclinical level. Our aim is to reach clinical level and use it in the treatment for HD patients.

 

Why did you apply to the aiForward program?

The AiForward program can give the opportunity to do image analysis faster and more unbiased.

Had you thought much about using AI in your research before applying to aiForward?

Yes, I had thought using AI in my research due to manual counting of the cells is time-consuming and quite often can be biased. To eliminate these issues it is necessary to use several researchers.

 

Tell us a little bit about your aiForward project.

My Aiforward project is based on the analysis of huntingtin inclusion, which can be found in in vivo models and in HD cases. There are several types of inclusions and the aiForward program can help to analyze them and separate based on the type of the aggregates. Nowadays there is no most convenient method to count these inclusions, however most of them are based on manual counting.

What are your expectations for the program?

Our expectations are that Aiforward can make our analysis faster and more accurate!

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

Interview with pathologists Helen Remotti MD and Ladan Fazlollahi MD

By News

“I believe there is a great need for AI-based image analysis techniques in pathology research, and eventually clinical diagnosis, as we need to move from subjective analysis of tissue components to more robust and objective quantitate methods.”

What do you do and what is your research focus?

We are academic pathologists with broad research interests involving liver/gastrointestinal/pancreatic benign and neoplastic diseases. Fatty liver disease has become an important area to study as it has become the leading cause of liver disease in United States and worldwide, and among the top reasons for liver transplantation. Here at Columbia University Medical Center, we receive a large number of liver biopsies from patients with alcoholic and non-alcoholic fatty liver disease. Therefore, we do have the resources and the opportunity to advance our study methods and understanding of this disease.

Why did you apply to the aiForward program?

As pathologists, we quantitate the degree of fat content in our liver biopsies as mild (5-33%), moderate (>33-66%) and severe (>66%) by estimating the percentage of fat under microscope. We also use the routine H&E slide and a trichrome-stained slide to assess the stage of fibrosis (Stage 1 to 4). There is some degree of inter-observer variation in estimating the percent fat and fibrosis stage. The data derived from this type of pathology analysis is used to direct patient care and even enrollment in clinical trials. Now with AI-based imaging techniques available, we are interested to see if these techniques could help us better estimate the fat content, differentiate the small fat-droplets from large fat-droplets and fibrosis stage in liver biopsies. In addition, AI based imaging can be used for quantitating other biomarkers.

Had you thought much about using AI in your research before applying to aiForward?

Yes. we have previously worked on multiple collaborative projects focused on pancreatic, liver, and colorectal carcinomas. We used AI-based image analysis techniques to evaluate and compare the immune cell density in tumor and stroma. I believe there is a great need for AI-based image analysis techniques in pathology research, and eventually clinical diagnosis, as we need to move from subjective analysis of tissue components to more robust and objective quantitate methods.

Tell us a little bit about your aiForward project.

We have a cohort of liver biopsies from patients with non-alcoholic steatohepatitis (NASH) and a cohort of liver biopsies from control patients (liver donors) with correlative Fibroscan liver stiffness measurements. FibroScan is a non-invasive ultrasound-based imaging technique used by hepatologists to indirectly assess the fat content and elasticity (corresponding to fibrosis) of the liver parenchyma. The aim of our study is to use the AI-image analysis platform developed by Aiforia for quantitative evaluation of fat and fibrosis in liver biopsies, and to correlate these findings with parameters obtained by routine pathology assessment and FibroScan.

What are your expectations for the program?

We would like to have quantitative data on percentage of fat and fibrosis in our cohort of liver biopsies within a short timeframe (less than 6 months). Quantitation of additional biomarkers is planned for future studies.

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

Interview with PhD student Ilmari Parkkinen

By News

All in all, we are very fortuitous to have been chosen for the program and look forward to making the most out of it.

What do you do and what is your research focus?

I am a PhD student, and I study Parkinson’s disease (PD), which is a debilitating age-associated neurodegenerative disease. The aim of my research is to elucidate why dopamine neurons degenerate in the disease and find potential new ways to treat it. Particularly, I am taking a closer look into the endoplasmic reticulum (ER) of these neurons with advanced imaging techniques and trying to find drug targets or compounds targeting the dopamine neuron ER. The ultimate goal of the research is to find disease-modifying (either halting or reversing) treatments for PD as the current ones only alleviate symptoms.

Why did you apply to the aiForward program?

With regards to preclinical PD studies, which our lab does, there is a lot of various morphometric analyses involved, especially cell counting. Cell counting is mostly done manually with laborious methods such as stereology, but AI can automate it and make it much easier and more reproducible. We wanted to create a faster and easier way to count cells and proteinaceous inclusions from cells, like Lewy Bodies which are found in PD patient’s brains. Because Aiforia offered a chance to create a custom algorithm just for this purpose, we decided to go for it and use it in our models.

Had you thought much about using AI in your research before applying to aiForward?

Yes, we had a few ideas, but it always came down to execution as using AI for your research requires a good amount of expertise and/or a good platform to get started. Since we had neither readily available, a collaboration was the way to go. We eventually found Aiforia, which has both, the expertise and the platform, and offers a hands down and convenient way to apply AI to imaging-related research needs.

Tell us a little bit about your aiForward project.

We are developing a CNN-based algorithm to count Lewy Bodies (LB) and Lewy Neurites (LN) from histological samples, in our case mainly brain sections. Ideally it would identify and locate different neurons (like the dopamine neurons which degenerate in PD) and the LBs and LNs, count how many there are, in which cells they are and also give different parametrics, such as size and distribution, for them.  

What are your expectations for the program?

Mainly we are expecting to do exciting and robust work that could lay the basis for our future studies to study the mechanisms underlying neurodegeneration. Also, at the end of the day, hopefully we will find a way to make the lives of neuropathologists and researchers doing preclinical studies on PD easier. The great thing is that we already have some promising preliminary data. However, it is still in its early phases to say how it compares to other counting methods or identification done by professional neuropathologists. All in all, we are very fortuitous to have been chosen for the program and look forward to making the most out of it.

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

aiForward project assessing PD-L1 in lung cancer nears completion

By News

“Immunotherapy has very beneficial effects, particularly in lung cancer.”

“When I first started, I did not know anything about AI,” describes Liesbeth Hondelink, MSc student at Leiden University Medical Centre in the Netherlands, when she applied to the aiForward program.

Her research focuses on PD-L1, a protein which has gained much popularity in a field that has been under the spotlight for some time now: cancer immunotherapy. Liesbeth and her lab are specifically assessing the programmed death-ligand 1, which is what PD-L1 stands for, in non-small cell lung cancer (NSCLC).

Prescribing the correct immunotherapy

“Immunotherapy has very beneficial effects, particularly in lung cancer,” explains Liesbeth. The percentage of tumor cells positive for PD-L1 in NSCLC has significant consequences on the choice of this expensive treatment since it predicts the effectiveness. The proportion of the protein must therefore be calculated to aid clinicians in prescribing the correct course of treatment.

This is calculated by specialized pathologists, an arduous task prone to inter- and intra-observer variability, as Liesbeth describes: “Pulmonary pathologists are finding it difficult to score PD-L1 on lung biopsies and disagree in 15-20% of all cases. This has been extensively assessed in literature.”

AI helps provide guidance

“Aiforia seemed like the most approachable, easy-to-use solution,” Liesbeth explains why she applied to aiForward. Using Aiforia Create, Liesbeth has already trained her own deep learning AI model to calculate PD-L1 in a speedy and consistent manner. “We annotated everything, about 60 slides, in just a few weeks and it was actually quite easy. Easier than I expected it to be and I expected it to take much longer,” she describes of the training process.

“Annotation Assistant really saves you a lot of time. It also brings up the more difficult areas that I myself would not have realized to annotate. It really made the algorithm better,” she adds, having used Aiforia’s new active learning tool to speed up her AI model creation.

Analysis under way

Training is done and now with the PD-L1 detecting AI model in her hands Liesbeth has already started the analysis with the support of the pulmonary pathologists in validating the results. “The Aiforia AI really does something that we could not do otherwise, and the results are looking good.”

Expecting to finish her aiForward project soon, this researcher transformed from not knowing anything about AI to creating her own deep learning AI model in just a matter of weeks.

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

ALS therapy research project launches with aiForward

By News

“Before I did not think that AI could do something like this, but then once I was shown with Aiforia, it became clear that this is what I want to use.”

Neurotrophic factors give neurons life. They are a group of molecules that support and enhance the growth and survival of neurons. The Voutilainen Lab studies regenerative neuroscience at the University of Helsinki’s Institute of Biotechnology. The lab’s mission is to identify treatments for diseases such as amyotrophic lateral sclerosis (ALS); a fatal neurodegenerative disease that leads to paralysis within just a few short years of diagnosis. There is currently no cure for the disease.

Two students in the lab, Liam Beckett and Jaan Korpikoski, are focusing their research on a novel protein with trophic factor properties, currently one of the most promising therapeutic candidates for ALS. The ultimate aim of this project is to take a drug candidate to clinical trial and to translate the results for the benefit of ALS patients. The aiForward program aims to accelerate this research.

Discovering new therapies for a debilitating disease

Liam came from the UK to brave the cold in Finland and start his PhD studies after years of experience in drug development research. Liam supervises Jaan, who is a master’s student. Both decided to focus their studies on ALS to help combat the devastating disease. “Therapy is very limited. When you get an ALS diagnosis you have a very short lifespan. It is a very debilitating disease and it is a huge burden for not only the patient but also for their care-giver” Jaan explains.

In order to assess the effects of the molecule in question their research relies heavily on the histological analysis of spinal cord samples from mouse models of the disease to identify and quantify motoneurons. With traditional methods this requires many hours of manual counting — a slow, tedious, and often unreliable methodology particularly as the neurons in question show a much variability. Therefore, when the aiForward program was introduced to Jaan, he was excited by the possibility of automating and standardizing this process.

Advancing research on promising drug candidate with AI

“Before I did not think that AI could do something like this, but then once I was shown with Aiforia, it became clear that this is what I want to use,” Jaan describes his reaction after being introduced to the platform. After his application was accepted Jaan started to develop the AI models himself through Aiforia Create, by training the AI to look for specific motoneurons. “We need an algorithm that counts them every single time, in the same way and then we can have reliable data. We need to decrease the variation as much as possible.”

The ALS aiForward project is in its early stages but both Liam and Jaan are already impressed with the technology and its clear benefits to their neuroscience research: “It is extremely difficult for researchers to identify these neurons by eye. Aiforia is especially helpful that it actually provides a standardized method. This is one of the biggest challenges for us in this research,” explains Liam. “It can also see things that most likely we cannot,” he adds.

Neither researcher had used AI before but within a few weeks’ time Jaan had begun to train his first AI model without the need for any coding: “I was surprised with Aiforia how easy it is to train the AI and how well it works. You just tell it what to look for,” he describes. Both are now looking forward to seeing this AI model in action and advancing their ALS research.

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

The first aiForward project commences

By News

“I have been thinking about these vessels, a lot. I am excited to finally be able to see them in a tissue context.”

Reidunn Edelmann, an MD currently doing postdoctoral research at the University of Bergen at the Centre for Cancer Biomarkers (CCBIO) in Norway, is one of the first successful aiForward applicants. Driven by a passion for clinical research, Reidunn completed his PhD in cell biology after her medical studies and is currently focusing on studying biomarkers and vessel characteristics in cancer. She is interested in changing paradigms over how biomarkers and vessels are studied, from simple “very much just focused on one marker,” as she explains, “to thinking about the wider, more functional picture.”

Deepening cancer research

The vasculature, the arrangement of blood vessels in a tumor, is a significant component of analyzing tumor microenvironment which in turn is considered an important prognostic factor in cancer. “Earlier analyses of tumor vasculature in clinical samples has generally been of “low-content” nature,” Reidunn states. Her mission is to take the analysis of this important biology to the next level by deepening the characterization of these vessels and to ultimately correlate them to patient outcome-related end-points as well as other biological traits of tumors in different cancer types.

In order to fulfill these goals, Reidunn and her lab created a complex, multiplexed fluorescent staining that potentially can visualize distinct vessel phenotypes, based on activation of an essential fate-decisive pathway. However, the next step proved to be difficult, “The samples are so heterogenous, the patterns are so different from each other even for the same staining quality,” Reidunn explains “Current and traditional methods to analyze these were too simplistic, not suitable for this complex staining.” She had the samples, she had the staining, but she did not have the necessary tools to enable analysis and discovery.

aiForward enables analysis

Reidunn came across Aiforia at a conference, intrigued by the possibilities of applying deep learning AI to her research she contacted one of Aiforia’s scientists to describe her problems: “Manually it would definitely not be possible to work on my research, it takes too much time with these complex samples.” She was encouraged to apply to the aiForward program, as she explained: “My motivation is that I really, really want to analyze complex, high-content samples. Deep learning AI can learn the different patterns of what the activation of one pathway looks like.”

Her project with aiForward has started off with uploading the samples of her cohort, 235 colorectal cancer samples of patients to the Aiforia Cloud Platform — the Aiforia scientists were actually able to build a custom solution to support the upload of her unique sample set. When asked how she feels about the next steps, which involves training the AI model to recognize her features of interest, Reidunn excitedly answers: “I think it will be so easy, Aiforia (software) does all the hard work, I don’t need to think about any algorithms. All I need to do is see my slides and look for my feature and annotate it and then the program is going to do the rest. That’s the amazing thing. I myself am not so into digital, technical stuff. I am very much biology focused. So, thanks to Aiforia, I can just focus on what I want to focus on, the biology.”

Applications to the aiForward program are accepted on a continuous basis.

Apply Now

3 reasons to use Deep Learning in Digital Pathology

By News

Deep learning (DL) is a subset of machine learning (ML) which falls under the category of artificial intelligence (AI). The formation of layers is not only pervasive in the nomenclature of these different types of computer intelligence but also in their architecture. DL is based on artificial neural networks, a layered and connected system of algorithms receiving and processing information.

Traditional ML can only be taught information, or say features in images, extracted and coded by a human. DL on the other hand is highly flexible and scalable, it can learn features by itself, making it an extremely powerful tool for image analysis. The Aiforia Platform enables easy access to deep learning and the creation of algorithms for the analysis of any histopathological image; allowing pathologists to speed up their work, make new discoveries, and work with consistency.

  1. Improves accuracy and speed

For the first time ever, deep learning AI models are able to mimic humans in learning to recognize complex visual features in image data. However, DL is faster, often more accurate and can therefore surpass human capability. It can be deployed in a huge variety of applications, from object quantification to tissue classification based on morphology, and give accurate, quantitative information from biological samples. Pathologists can therefore now automate manual and time-consuming image analysis work.

Time is given back to focus on more important tasks: decision-making, collaborating with peers, and saving time away from being hunched over a microscope for hours on end. With the convenience of a portable tablet or other device the pathologist can now automate analysis and review the data and examine the specimen, anywhere, any time.

In one application, an Aiforia deep learning AI model detected and counted all dopaminergic neurons in rodent brain substantia nigra in 5 seconds, compared to the normal 45 minutes with traditional methods. Furthermore, the speed with which this analysis was completed, did not hinder the accuracy, as this was comparable to the StereoInvestigator used previously in the lab.

  1. Enables new discoveries

Deep learning not only can outperform humans in accuracy and speed, it can also find what the human eye sometimes cannot see on its own. Features that are either too small, heterogenous in their expression, or found in a huge quantity spread over a large area are easily recognizable by AI. The process of discovery can be enhanced, especially when considering the development of novel drug molecules, as deep learning excels at detecting subtle differences between study groups.

The smallest of changes, not possible to visualize manually, can be detected by the AI models. Furthermore, you can train them with external ground truths not readily identifiable from the images provided, such as response to treatment. Predictive models can be trained with outcome data. These then learn to extract and visualize the features associated with the outcome.

In one research project AI models were developed with Aiforia Create to enable a pathologist to quantify the number of liver cells positive for a specific biomarker indicating bile duct injury. Without Aiforia this task would have been nearly impossible, requiring multiple, highly-trained pathologists all detecting and quantifying cells in the exact same, uniform manner for several hours a day for many weeks. The increased quantitative and novel data produced by deep learning AI allows pathologists to better assess and determine diagnostic and prognostic indicators and criteria, leading healthcare many steps closer to providing personalized medicine.

  1. Completes tasks consistently

Inter- and intra-observer subjectivity is a critical issue in image analysis across diagnostic settings, many research fields, and many sample types. Whether a specific feature is scored, can vary from one person to another, and even within the same pathologist from one day to the next. Deep learning AI however doesn’t have bad days, it doesn’t get tired, its performance and observation stays consistent. It does not stray away from what it has been taught to do or find. The algorithms classify results and solve problems with concordance, always according to the ground truth they were given.

In one such application, Aiforia was deployed by pathologists at the pharmaceutical company Sanofi to count large quantities of Th+ neurons. Over 150,000 objects were detected in the samples for this one study. Not only was the process speedier than with traditional, manual methods but absolute consistency was achieved. The AI models analyzed and quantified each slide and section with the exact same ground truth in mind. Find out more about how Aiforia enabled this preclinical research.

Conclusion

Aiforia is a software platform allowing users to create their own deep learning AI models to automate the analysis of a variety of tasks in a variety of histopathological images, without any coding or hardware needed. Deep learning AI should be viewed as an assistant to the pathologist, not a replacement. A tireless, consistent, speedy assistant, here to help you discover more. Find out more about the Aiforia Platform here.

This article originally appeared on www.aiforia.com, read the original article here: https://www.aiforia.com/blog/3-reasons-to-use-deep-learning-in-digital-pathology/