Artificial Intelligence

eusomii-annual-meeting-2019-senza-indirizzo

QUIBIM HOSTING EUSOMII ANNUAL MEETING 2019

Next 18th and 19th October 2019, Valencia will host the Annual Meeting of the EUSOMII society (AMI2019). The society has planned a two days congress combining educational and scientific sessions in the field of medical imaging informatics.

In this edition, QUIBIM will participate with 2 lectures, 3 oral communications, 7 posters and a QUIBIM Precision exhibit space for all those interested in our artificial intelligence algorithms and imaging biomarkers solutions. If you want to book a demo, do not hesitate and book a timeslot!

Book a DEMO

As a board member of EUSOMII, it is an honor to bring to Valencia, our city, such an interesting program. Holding this meeting in an innovative environment like the Hospital Universitario y Politécnico La Fe de Valencia is a sign of the change that healthcare is experiencing nowadays” explained Ángel Alberich-Bayarri, CEO & Founder of QUIBIM and Chair Industry and Startup Committee of EUSOMII.

Check out our participations at AMI 2019:

FRIDAY, October 18th 2019

  • Keynote Lecture I – Imaging biomarkers and radiomics: source of big data for AI – Dr. Luis Martí-Bonmatí

SATURDAY, October 19th, 2019

  • Didactic Lecture II-III – Imaging Biomarkers – Dr. Angel Alberich-Bayarri
  • Oral Communication – Image analysis using an intercontinental infrastructure for the deployment of
    trustworthy cloud services: the ATMOSPHERE project. Authors: Ignacio Blanquer, Eduardo Camacho-Ramos, Andrey Brito, Ana Jimenez-Pastor, Christof Fetzer, Altigran da Silva, Amanda Calatrava, Fabio García-Castro, Ángel Alberich- Bayarri, Franciso Brasileiro.
  • Oral Communication – Quantification and evaluation of pre-post exercise femoral cartilage thickness and T2
    changes in ultramarathon athletes. Authors: Fabio García-Castro, Jordi Catalá March, Daniel Brotons Cuixat, Miquel Llobet Llambrich, Eduard Sánchez Osorio, Ángel Alberich-Bayarri.
  • Oral Communication – Automated Lung Segmentation in Chest Radiographs using Deeply Supervised Convolutional Neural Networks Trained by means of a Database Augmented with a Generative Adversarial
    Authors: Rafael López.

POSTERS:

  1. Automatic cartilage segmentation in 3D T2w high resolution MR using a Deeply
    Supervised Multi-Planar Convolutional Neural Network. Authors: Ana Jimenez-Pastor, Fabio García-Castro, Ángel Alberich-Bayarri, Luis Marti- Bonmati.
  2. Automatic quantification of white hyperintensities in a healthy aging cohort using
    Convolutional Neural Networks. Authors: Ana Jimenez-Pastor, Eduardo Camacho-Ramos, Ángel Alberich-Bayarri, Carles Biarnes, Josep Garre, Joan Carles Vilanova, Rafel Ramos, Reinald Pamplona, Salvador Pedraza, Josep Puig.
  3. Adaptation of TLAP-certified radiological structured reports to be used in a cloud
    platform environment. Authors: Fernando Bacha-Villamide, Eduardo Camacho-Ramos, Alejandro Mañas-Garcia, Luis Martí-Bonmatí, Angel Alberich-Bayarri.
  4. Development and validation of an inter and intra-sequence registration algorithm in
    multiparametric prostate resonance imaging. Authors: Matías Fernández, Mar Roca-Sogorb, Fabio García-Castro, Raúl Yébana, María Asunción Torregrosa, Leonardo Bittencourt, Margarita García Fontes, Paula Pelechano, Luis Martí- Bonmatí, Ángel Alberich-Bayarri.
  5. Computer aided diagnosis for Rheumatic Heart Disease by AI applied to features
    extraction from echocardiography. Authors: Eduardo Camacho-Ramos, Ana Jimenez-Pastor, Ignacio Blanquer, Fabio García- Castro, Ángel Alberich-Bayarri.
  6. Outcome prediction after acute stroke through functional magnetic resonance imaging. Authors: Eduardo Camacho-Ramos, Ana Jimenez-Pastor, Carles Biarnes, Ángel Alberich-Bayarri, Salvador Pedraza, Josep Puig.
  7. Implementation of an interactive radiological structured report management system
    with AI annotation capabilities. Authors: Alejandro Mañas-Garcia, Eduardo Camacho-Ramos, Ismael Gonzalez, Fernando Bacha, Ángel Alberich-Bayarri, Luis Marti-Bonmati.

Our Quibimers are already heating engines for AMI2019. We look forward to meeting you!

QUIBIM_Curso IA Banner

CURSO PROPIO: Introducción a la Inteligencia Artificial aplicada a la Imagen Médica

Tenemos el placer de anunciar la primera edición de nuestro curso: Introducción a la Inteligencia Artificial aplicada a la Imagen Médica. 

Se trata del primer curso teórico-práctico centrado en dar conocer más a fondo algunos temas de máxima actualidad relacionados con la creación de algoritmos básicos de preproceso de las imágenes médicas y biomarcadores de imagen. Además, se practicará el desarrollo de algoritmos de inteligencia artificial basados en redes neuronales convolucionales aplicados a las imágenes médicas.

Dirigido a estudiantes de grado y máster de ingeniería biomédica, telecomunicaciones, informática, ciencia de datos, matemáticas, así como de otras carreras técnicas. Investigadores interesados en la imagen médica, la inteligencia artificial y los biomarcadores de imagen.

Puedes realizar la inscripción en las siguientes modalidades:

Recurso 4

PROGRAMA – Horario de 17 a 20 hrs.

DÍA 1 – Lunes 18 de noviembre de 2019 (3 horas)

  • Introducción al curso: la imagen médica desde Valencia al mundo.
  • Modalidades de adquisición:
  • No ionizantes (resonancia magnética y ultrasonidos)
  • Ionizantes (rayos X, tomografía computarizada y medicina nuclear) e imagen híbrida.

DÍA 2 – Martes 19 de noviembre de 2019 (3 horas)

  • Estándares en imagen médica: DICOM y otros.
  • Ejercicio-Notebook: Manipulación de formatos de imagen médica.
  • Repaso de conceptos.

DÍA 3 – Miércoles 20 de noviembre de 2019 (3 horas)

  • Procesamiento de imágenes médicas.
  • Biomarcadores de imagen: ¿Qué son?
  • Biomarcadores estructurales
  • Biomarcadores funcionales
  • Ejercicio-Notebook: Análisis de imágenes.

DÍA 4 – Jueves 21 de noviembre de 2019 (3 horas)

  • Introducción al Machine Learning
  • Ejercicio-Notebook: Inteligencia artificial aplicada: Clasificación con Redes Neuronales Convolucionales

FECHA: Del 18 de noviembre al 21 de noviembre de 2019

LUGAR: (COITCV) Col·legi Oficial d’Enginyers de Telecomunicació de la Comunitat Valenciana. Avinguda de Jacinto Benavente, 12. 46005, Valencia, España

¿Te interesa? Puedes realizar la inscripción en el siguiente enlace:

InscribemeMás info: PROGRAMA 

ALZ WORLD DAY

Imaging Biomarkers in Alzheimer

Alzheimer’s disease (AD) is a neurodegenerative disorder considered the most common cause of dementia worldwide. Dementia is a general term for the loss of cognitive functions –memory, thinking or reasoning- and behavioural abilities that affects the daily life of a person. According to the World Alzheimer Report, about 50 million people worldwide suffer from dementia. This number is estimated to be almost doubled every 20 years, reaching 131.5 million patients in 2050. AD is deemed the main form of dementia and contributes to 60-70% of the cases.

In this disorder, the connections between the nerve cells that make up the brain are affected, causing the death of these cells and the loss of brain tissue. In AD, abnormal levels of beta-amyloid and tau proteins are found in the brain. Evidence suggests that a complex interaction between these proteins is the main responsible of Alzheimer’s brain changes. Depending on the brain regions affected, different alterations may be presented in the patient. Some of the first regions to be altered are the entorhinal cortex and hippocampus, which are closely related to memory performance.

AD is a progressive disorder in which more parts of the brain are damaged over time. As this happens, more dementia’s symptoms are developed and the patient’s condition gradually worsens. In the early stages, people often present a reduced ability to take and remember new information; and may be accompanied by word-finding problems, vision or spatial issues, or impairment in reasoning or judgement. As AD progresses, patients suffer an increased memory and cognitive loss. They can experiment difficulties in recognizing family and friends, loss the ability to have conversations or be unable to respond to their environment. In severe cases of AD, brain tissue shrinks significantly, making people unable to communicate and completely dependent on others.

There are no effective treatment options for AD patients able to detain its progression. Nevertheless, symptoms can mitigate by means of medication and a healthy lifestyle. Some risk factors, such as age and genetics, are out of control; while others can be overcome to take care of brain-health. According to the Alzheimer’s Research & Prevention Foundation, regular physical exercise can reduce the risk of developing AD up to 50%. Others such as social engagement, healthy diet, mental work and stimulation, good sleep and stress management has proven to fortify the brain and reduce the risk of suffering any form of dementia.

As shown, prevention is the most effective way to fight this devastating disease. An early diagnosis is fundamental for this purpose. Nowadays, this is performed mainly by means of questions to the patients, blood/urine tests, and memory, attention or problem-solving exercises. Brain scans, such as computed tomography (CT), positron emission tomography (PET), or magnetic resonance imaging (MRI) are also key elements in diagnosis because they allow to detect the presence of abnormal concentrations in proteins and atrophy.

In this way, our team has developed automatic brain tools for the detection and assessment of Alzheimer in a subject. Our Brain Atrophy suite and Voxel-Based Morphometry analysis modules are designed to evaluate the shrinkage of the brain by comparing its volume and morphology with a set of healthy subjects (paired in age and gender).

two examples of Brain Atrophy- Hippocampal Asymmetry analysis reports

Above there are two examples of Brain Atrophy: Hippocampal Asymmetry analysis reports over a healthy and over an Alzheimer’s brain. This analysis provides a view of the brain focused on hippocampus’s status (volume and level of asymmetry) framed in a comparison with a set of healthy people. Taking a look at them, relevant differences in the Alzheimer’s brain come to light: the left hippocampus volume falls below the normal ranges (red region in the table), while the left-right hippocampus asymmetry is also presented out of normal values.

Other important tool to assess brain atrophy is our Voxel-Based Morphometry analysis module, which gives volumes and statistical scores comparing morphology differences between the pathological brain and the previous set of healthy subjects.

QuibimStructuredReport_Voxel-Based Morphometry-01

The analysis above has been performed over the same Alzheimer’s brain than the previous Hippocampal Asymmetry evaluation. Both reports highlight alterations in the hippocampus, which is one of the first affections in AD.

Other tools we have included in the Brain Atrophy suite, such as global brain screening, frontal-temporal dementia and motor cortex evaluation can be also performed to obtain a more enlightening sight of the brain.  The combination of these brain assessment tools with traditional AD evaluation methods can lead to a better understanding and an earlier diagnosis of this terrible pathology.

2019-06-26

QUIBIM to develop platform in leading research project to fight pediatric cancer

QUIBIM is helping to advance knowledge of the most lethal pediatric tumors through EU-funded program PRIMAGE, which exploits precision information from medical imaging to establish tumor prognosis, and expected treatment response using radiomics, imaging biomarkers and artificial intelligence (AI).

Pediatric cancer is a rare disease, but treatment remains challenging. Improving knowledge is key to adequately plan therapy and boost survival, and the latest AI techniques have the potential to harness unprecedented information from medical images.

A few months ago, the European Commission funded the PRIMAGE (1) project with over €10M, to help identify the most efficient treatment and a tumor’s main characteristics without the need for biopsy, by using computational processing of medical images on the cloud.

The PRIMAGE consortium will create a bank of images obtained through AI, using an open cloud-based platform to support decision-making in the clinical management of Neuroblastoma (NB), the most frequent solid cancer of early childhood, and Diffuse Intrinsic Pontine Glioma (DIPG), the leading cause of brain tumor-related death in children. The PRIMAGE platform will implement the latest advancement of in-silico imaging biomarkers and modeling of tumor growth towards a personalized diagnosis, prognosis and therapies follow-up.
The project involves 16 European partners, including internationally recognized institutions, and four leading industrial partners, including Spanish biotechnology company QUIBIM, all working under the aegis of the Imaging Biomedical Research Group (GIBI230) based in La Fe Hospital, Valencia.

181220_IISLAFE PRIMAGE kick offprimage-logo-transparent
Sharing high-end knowledge of AI tools

QUIBIM is responsible for the central task of developing the PRIMAGE platform’s architecture, adaptation and design. The company recently obtained CE mark for its Chest X-Ray Classification AI-Tool and its imaging biomarker analysis algorithms, zero footprint DICOM viewer and platform within the QUIBIM Precision platform.

QUIBIM researchers are now bringing their expertise in medical image post processing and management to PRIMAGE, by passing on their knowledge of clinical trials design and validation, imaging biomarkers extraction and validation, radiomics, data clustering and visualization, and development of AI-fueled tools, such as organ segmentation models.
“Much work remains to be done to improve our knowledge of pediatric brain cancer. NB and DIPG have a complex therapeutic approach and we need proper tools to improve survival. Extracting quantitative information from medical images with AI can help visualize tumor growth with extreme precision, and help to tailor therapy to each individual patient,” Ángel Alberich-Bayarri  said.

QUIBIM’s input will also help to define the methodologies and standards to be used in the different development areas, to facilitate interoperability between the platform ́s modules and for future interoperability with their cloud-based platforms for functionality add-ons.

Transferrable knowledge to other cancers

Cancer has a very low incidence among children and experts estimate that 500,000 EU citizens will be pediatric cancer survivors by 2020. Nonetheless, cancer remains the first cause of non-traumatic death among children.

Neuroblastoma is the most common extracraneal tumor in children, representing 8-10% of all pediatric cancers. In Europe, 35,000 new cases are diagnosed each year, 1,000 in Spain alone.

Diffuse Intrinsic Pontine Glioma is a very rare disease in childhood and is associated with low survival (10%), despite many existing treatments and on-going research. Treatment is not curative, only palliative, i.e. radiotherapy to improve the patient’s life. 16 new cases are diagnosed each year in Spain, accounting for 2.5% of oncological pediatric patients and 13% of pediatric tumors of the central nervous system.

Because of the peculiarities of computational approximation in these two types of tumors that are proper to childhood, investigation done in that area will also be applicable to other types of tumors. Because it will gather considerable scientific effort, PRIMAGE should also help advance research on other types of cancer.
(1): PRedictive In-silico Multiscale Analytics to support cancer personalized diaGnosis and prognosis, Empowered by imaging biomarkers (PRIMAGE)

MS WORLD DAY_QUIBIM

IMAGING BIOMARKERS IN MULTIPLE SCLEROSIS

Pain, depression, slurred speech and feeling of numbness, tingling, or weakness. These are just some symptoms of Multiple Sclerosis (MS) disease, a long-term condition that affects the brain and spinal cord.

In MS, the immune system confuses myelin with a foreign body and attacks it. The loss of this protective sheath that covers the nerve fibers will disrupt the messages travelling between the brain and the body. These messages may be slowed down, interrupted, or may not occur at all.

Eventually, the person sees affected the ability of controlling their own actions. The signs and symptoms of MS may vary greatly depending on the stage of the disease and the location of the affected nerve fibers. Movement affections, vision problems, altered speech and dizziness are common consequences of this condition.

It is the most widespread neurological disorder of young adults globally. The disease can be developed at any age, but its main incidence appears in the range of 20-50 years old. The National Multiple Sclerosis Society estimates that near 2.3 million people are living with this disease around the world. It also calculates that 1 million of them are placed in the United States, where 200 new cases are diagnosed every week.

Together with blood tests, medical history and neurologic exams, imaging scans have also proven to be a key element for the diagnosis of MS, concretely the Magnetic Resonance Imaging (MRI) is the reference diagnostic technique for the identification of lesions in MS.

Damaged white matter has a prolonged T2 relaxation time due to increased tissue water content and to degradation of the myelin, being well depicted on MRI and concretely on Fluid Attenuated Inversion Recovery (FLAIR) images. In this MR-sequence, MS lesions are seen as white matter hyperintensities (WMH). Nowadays, manual segmentation of WMH areas is still the gold standard to quantify the total lesion volume and to know the number of lesions in the brain. However, this methodology turns MS patient’s diagnosis and follow-up in a cumbersome and time-consuming task with high intra- and inter- observer variabilities.

Zero-click tools based on Artificial Intelligence (AI) and, more concretely, Convolutional Neural Networks (CNN) can be used to automatically segment WMH on FLAIR images in a few minutes. Novel designed architectures are composed of an ensemble of CNNs built on standard convolutional, dilated and residual layers.

Multiple Sclerosis_QUIBIM

These tools are capable of fine segmentation of the lesion avoiding the physiological WMH as the ependymal layer. Physicians can obtain quantitative information that helps them to achieve a more accurate and earlier diagnosis, thus reducing the workload and improving the time-efficiency while enhancing patient assessment.

What information does it provide?

Once WMH are segmented, relevant lesion statistics are quantified: lesion number, total lesion volume, dominant lesion volume, dissemination, or entropy among others. All this information can be summarized in a structured report along with the most characteristic slices. These processes will easily assist physicians in the diagnosis of MS patients not in the future, but now.

QuibimStructuredReport_White matter lesions

AUTHORS:

Eduardo Camacho Ramos.

Ana Jiménez Pastor.

QUIBIM_chest_xray_classifier_logo3

QUIBIM’s AI-FUELED CHEST X-RAY CLASSIFIER GETS CE MARK

  • Chest X-ray classifier is added as a new CE cleared tool within QUIBIM Precision platform, which already received CE mark class IIa certification earlier in 2019.

  • QUIBIM’s Chest X-Ray Classification AI-Tool dynamically learns using new images, meaning the system is continuously evolving and improving over time.

Valencia, Spain – 27th May 2019. Spanish healthcare AI company QUIBIM today announced that its AI-powered Chest X-Ray Classification tool has received CE certification. The company already obtained the class IIa CE mark earlier this year for the imaging biomarker analysis algorithms, the zero footprint DICOM viewer and the platform within the QUIBIM Precision platform, becoming the first Spanish firm to ever receive the clearance.

QUIBIM applies machine learning and image processing techniques to extract imaging biomarkers from medical images in order to assist radiologists and physicians in daily practice. With its AI-based chest x-ray classifier, QUIBIM helps to detect pathological findings that could go unreported due to the heavy workload of radiology departments.

QuibimStructuredReport_Chest X Ray Classifier

“Our tool was designed to prioritize unreported, potentially pathological radiographs, to help radiologists be more efficient by focusing their efforts on studies that are more likely to have pathologies. We are sure this will have an impact in big healthcare systems,” Angel Alberich-Bayarri, QUIBIM CEO and founder, explained.

The solution makes use of a novel architecture based on referee networks combined with Convolutional Neural Networks that have been trained with a database of more than 500,000 images, to calculate the final probability of the X-Ray of being abnormal. Afterwards, the probabilities are used to estimate the presence of pathologies in chest X-Rays.

Because of this Artificial Intelligence methodology, the classifier understands the visual patterns that are most indicative of the different pathologies using the knowledge extracted from the large dataset of radiographs used to train the networks. “QUIBIM’s Chest X-Ray Classification Tool is able to learn further using new images, which means that this system is continually improving and evolving with time,” Rafael López, Artificial Intelligence Engineer at QUIBIM, said.

QUIBIM’S Chest X-Ray Analysis Tool is already available at QUIBIM Precision®, accessible through the cloud with just a few clicks or it can be fully integrated in the radiology department’s workflow as a local solution for seamless interpretation of chest X-Rays.

Breast cancer QUIBIM

QUIBIM uses AI to boost
breast cancer research

As the first ESMO Breast Cancer Congress unfolded 2-4 May in Berlin, Germany, QUIBIM CEO highlighted the role played by artificial intelligence (AI) in breast cancer research.

The European Society of Medical Oncology focused on the progress in treatment options and improved outcomes for breast cancer (BC) patients during its first meeting dedicated to the topic (¹).

One of the key messages of the conference was that therapeutic innovations should go hand in hand with a multidisciplinary, fully integrated approach to patient care, a scenario in which AI can increasingly help make a difference, according to QUIBIM CEO & founder Angel Alberich-Bayarri.

Breast cancer_QUIBIM

QUIBIM breast cancer structured report

AI, and in particular image quantification, can help significantly advance knowledge of this multi-faceted disease, by enabling earlier and better detection,” he said.

From the Pacific to the rest of the world

Recently QUIBIM partnered with the AI Precision Health Institute (AI-PHI) at the University of Hawai‘i Cancer Center in Honolulu, to manage, store and quantitatively analyze medical images and algorithms in breast cancer research.

Using the QUIBIM Precision® image analysis platform, AI-PHI researchers will create large scale imaging repositories with automated extraction of imaging biomarkers to characterize patients’ status. The solution will first be used as the central repository for the mammography studies conducted in the Pacific and will include mammograms from over 5 million women.

The University of Hawai‘i Cancer Center is one of 69 NCI designated cancer centers in the United States and the only one in Hawai‘i and all of the Pacific Islands. It is the only institution that uses AI to analyze medical images to assess health and predict risk of disease in the region.

The project is expected to grow exponentially as its principal investigator Dr. John Shepherd is keen on inviting selected reference centers all across the world to join the network. “It is a very interesting and pioneering work for QUIBIM, and our first big cooperation in the breast cancer setting,” Alberich said.

Seamless integration into clinical workflow

The QUIBIM Precision® image analysis platform enables to incorporate AI algorithms regardless of the institution where it is deployed, to facilitate integration into workflow.

“Researchers can program their own algorithm and code, and add it as a plugin to the platform. This feature, combined with the advanced storage and multi-user annotation capabilities of the platform, allows for a wide adoption within research groups and institutions working with AI. Time for deployment of algorithms in the real world is shortened by 25%, since you have the whole AI pipeline in just one place,” he explained.

The solution, which includes not only quantitative image analysis but also structured reporting capabilities, can be integrated into the radiology workflow and add value to this service. It can for instance be used to integrate AI algorithms to detect cancer without the need of radiologists in a first-read.

Earlier this year the platform received CE Mark certification as class IIa Medical Device, including the imaging biomarker analysis algorithms, the zero footprint DICOM viewer and the platform hosting these components and medical imaging data. The platform is now available for commercial deployment in European hospitals and diagnostic imaging centers.

In addition to the platform, the certification includes 15 algorithms in the five key areas of interest of the company: oncology, neurology, musculoskeletal, liver and lung

Also included in the CE Mark certification is the DataMiner, which was designed in collaboration with the team of Prof. Daniel Keim from Konstanz University to provide advanced visual analytics of large databases of patients for population health management and scientific exploitation. The DICOM web viewer that enables the user to visualize the images of a sequence, draw ROIs or apply filters is also included in the certification.

QUIBIM applies machine learning to develop tools for imaging data quantification, to accelerate image reconstruction, segmentation, detection and data mining. Beyond the breast, the company has created AI algorithms to help detect changes produced by brain and prostate cancer, but also other diseases in the hematological scenario such as non-Hodgkin’s Lymphoma.

(¹): https://www.esmo.org/Conferences/ESMO-Breast-Cancer-2019?utm_campaign=PRESS%20-%20Breast&utm_source=hs_email&utm_medium=email&utm_content=72075686&_hsenc=p2ANqtz-9pH5mSHFhwzXnTGp6oSizohrD75uXccBxo1woBVGUFhJl3K5y_p4Ms4Hoa3LcwRJ_KvILSmOWhsrZxTZSWLwumEXOnOw&_hsmi=72075686

D3YI165WAAUCunr

Comprehensive solutions will boost AI use in medical imaging

Integrating artificial intelligence (AI) into the radiologist’s workflow is still challenging, but all-inclusive solutions can help unfold algorithms’ power, QUIBIM CEO and founder Angel Alberich-Bayarri explained during the ESR AI Premium Event, which was hosted by the European Society of Radiology (ESR) and the European School of Radiology (ESOR) earlier this month in Barcelona.

Medical imaging AI is a bubbling field but very few solutions are used in daily practice today, Alberich told delegates during the busy meeting, which gathered top researchers in medical imaging AI and thousands of online attendees. “We have a lot of research, AI algorithms and start-ups, but few are really embedded in the radiology workflow,” he said.

A main obstacle to AI integration is a lack of knowledge of utilities. For most imaging biomarkers, the real application relationship with clinical endpoints on a large scale – diagnostic, prognostic and treatment response – remains unknown. Clinicians don’t want to integrate biomarkers that have not been validated, but if they don’t gather information massively and try to understand how biomarkers relate to the disease, they will never help advance healthcare, Alberich explained.

“We have to change our minds and not wait for biomarkers to be validated before they can be extracted on a daily basis. Similarly to genomics: we have to do sequencing and study diseases to detect unknown mutations by ourselves,” he said.

The lack of annotated data is another challenge. Radiology reports are not filled with a focus on data annotation, but on descriptive language that needs to be processed by natural language processing (NLP). However they would be an ideal source of knowledge, and not just for the clinicians.

One-stop-shop platform using biomarkers

QUIBIM was created in 2015 as a spinoff company of La Fe Hospital in Valencia, Spain, to help radiologists make the most of AI in their everyday practice, by providing a one-stop-shop solution using imaging biomarkers and powerful algorithms.

“Offering a single solution embedded in workflows is key because radiologists will not buy all start-up micro developments but the best platform,” he said.

The very name of the company is an acronym and stands for Quantitative Imaging Biomarkers in Medicine. Imaging biomarkers enable to measure everything happening in the body to extract parameters that can provide information on the tissue of the lesion type beyond classification. Fuelled by AI, these biomarkers can deliver unprecedented information on disease.

“Many different imaging pipelines have dramatically changed thanks to AI integration. For example, we used to have lots of problems doing segmentation with traditional algorithms in tissues and organs such as the liver in MRI. Thanks to AI, segmentation has improved and with it our knowledge of liver disease,” he said.

A prerequisite is to integrate data mining solutions to make radiomics easy to everyone. This means it must be embedded in the same platform, as current solutions are not able to treat this amount of quantitative data from patient cohorts. “A lot of parameters are quantified in daily routine, but there is still no way to store and process them massively. Our current PACS systems are simply not prepared for quantitative data,” Alberich said.

There is a lot of sense in working within a structured report (SR), as it enables to annotate data that will further advance research. Prospectively the studies can be very well annotated if AI imaging biomarkers are integrated in the fields of the SR. Working with the SR would tremendously facilitate communication between clinicians and with patients.

“We are building the radiology report of the future, only we’re doing it now. Patients do not understand radiology reports, so we have to change the way we communicate. We are very much aligned with the standardized way blood test findings are reported, everyone understands whether findings are in or out of range and if there is any abnormality. We have to be more intuitive in our communication,” Alberich said.

QUIBIM is developing quantitative, one-page long structured reports that are actionable, quantitative and automated. The reports are designed with KOLs in each speciality, to make sure they reflect the reality of clinical practice.

Clinical input and powerful technology

Cooperation with clinicians is a key axis for the company, which uses a stepwise model to create new solutions with clinicians in the loop.

The company develops highly performing AI algorithms using unprecedented network architectures, for instance, multiscale convolutional networks ensemble in brain MR, 2.5D network in liver MR and referee network in chest x-ray, and hundreds of thousands of annotated images that have been acquired through cooperation with researchers and university hospitals.

QUIBIM has launched algorithms in different product areas to provide a head to toe imaging biomarker solution; in neuro for Alzheimer, ALS, MS, stroke and cerebrovascular events; in MSK for articulations; in chest for screening, COPD and fibrosis; and in breast for screening. More products will soon be released that will focus on other body areas.

More than 60 hospitals worldwide currently use QUIBIM tools, most of which have received CE marks and/or are FDA pending. The company has notably developed the new ESOR teaching platform and has supplied its QUIBIM Precision® image analysis platform to the AI Precision Health Institute at the University of Hawai‘i Cancer Center in Honolulu. QUIBIM has received €3.5m funding ever since its creation and has offices in Spain and the U.S.