Part IV: The Future of Medical Scanning

The future of medical scanning is prognostic; hyperpolarised MRI scans are very promising since they can identify trace levels of biomarkers indicating future disease. Carbon-13-enriched glucose can act as a tracer to detect abnormal metabolism characteristic of burgeoning Type-2 diabetes. Para-hydrogenated bicarbonate ions can be injected to examine microenvironment pH levels suggestive  chronic inflammation.

Preemptive molecular scanning has the potential to form a prognosis before symptoms even appear – saving patients from years of suffering.

Another new avenue of development in the medical scanning sector that is gaining traction is the utilisation of artificial intelligence (AI) in analysing scans. Currently, a large wealth of information can be yielded from a single patient visit; however, trawling through vast amounts of data makes the diagnostic process time-consuming and labour intensive – increasing the risk of misjudgement and oversight.

Additionally, at the moment, a large proportion of valuable scan data is eventually discarded. This presents a significant waste of Big Data which could be repurposed for training the neural networks of AI systems to recognise subtle patterns in a scan, hence supporting faster and more reliable diagnoses.

Within the last decades, imaging techniques are being rapidly perfected and fine tuned even for extremely niche requirements. The rapid emergence of new technology and the tremendous advancement in medical scanning is epitomised by the very recent development of the highly-sensitive wearable MRI scanner which is adept at imaging the intricate interplay of bones, tendons and ligaments in motion; the image produced by this cutting-edge technology is reminiscent of the very first radiograph featuring the hand of Röntgen’s wife.

The juxtaposition of the two scans (taken just over a century apart) highlights the major strides made in the field of medical scanning.

In these scans, I do not see my death, as Röntgen’s wife did, but rather, the limitless potential for life-saving scanning technology.

---

---

The figures quoted in this essay have been meticulously researched and documented. Below is the bibliography for this section of my essay:

“Super-Cool Imaging Technique Identifies Aggressive Tumors.” National Institute of
Biomedical Imaging and Bioengineering, National Institutes of Health, https://www.nibib.nih.gov/news-events/newsroom/super-cool-imaging-technique identifies-agressive-tumors [Last Accessed: 03 Mar 2019]

Zhang, Bei, Daniel K. Sodickson, and Martijn A. Cloos. “A high-impedance detector-array glove for magnetic resonance imaging of the hand.” Nature Biomedical Engineering 2 (2018): 570-577. Nature. Print. [Last Accessed: 25 Feb 2019]

Image Credit:

Zhang, Bei, Daniel K. Sodickson, and Martijn A. Cloos. MR image from wearable glove
coil design. Nature Biomedical Engineering, Nature, May 2018. Web.
https://www.researchgate.net/figure/Wearable-glove-coil-design-and-images-a-Photos-of-the-glove-coil-with-the-hand-stretched_fig4_324952145

Röntgen, Wilhelm. “Hand mit Ringen (Hand with Rings).” Wikipedia, Wikimedia
Foundation Inc., 22 December 1895

Part III: Nuclear Magnetic Resonance

As opposed to x-ray and ultrasound scanning devices, MRI uses the body’s innate magnetic properties to produce scans. Magnetic resonance imaging takes advantage of the high abundance of hydrogen nuclei (protons) in soft tissues of the body (which contain water and fat) and the fact that, often, diseases are manifested as an increase in water content as a natural consequence of inflammation.

Every hydrogen proton in the body has a spin and, hence, a small magnetic moment (1). An MRI machine applies a very strong magnetic field across the body to align all the hydrogen protons with the field (2). A radiofrequency pulse is then introduced which forces the hydrogen protons to realign against the magnetic field (3). Once the pulse is removed, the hydrogen protons return to their original alignment and release electromagnetic energy as MR signals which can be detected (4).

Tissue can be analysed and differentiated based on how much energy its constituent hydrogen protons release and how long it takes for them to return to their original  alignment (the relaxation time). Since the development of the first MRI scanner, several variants of the underlying principle have emerged to address niche situations.

BOLD fMRI
Blood-oxygen-level dependent functional MRI (BOLD fMRI) works on the principle that an increase in neuronal activity is always accompanied with an increase in oxygenated blood flow in local capillaries as the demand for oxygen increases.

The haemoglobin which delivers oxygen to the activated neurons is paramagnetic when deoxygenated (T-state) but diamagnetic when oxygenated (R-state); this unique magnetic property (caused by its different quaternary structures) results in slightly different electromagnetic emissions from the hydrogen protons within the haemoglobin depending on the level of oxygenation.

The MR signal can be analysed to infer a plethora of information – allowing clinicians to monitor the level of brain activity in a patient suffering from neurodegeneration and enabling researchers to discover the regions of the brain responsible for both instinctive reflexes and complex contemplation.

With its extreme versatility and minimal health risks, MRI arguably forms the heart of modern radiography – however, for certain patients with underlying ailments, the very scanning modality that is renowned for its safety can be deadly.

Safer Contrast Agents
Contrast agents are frequently used in MRI scans to alter the relaxation time of the hydrogen protons through their interaction with strongly paramagnetic gadolinium ions in the dye – this changes the MR signal intensities and increases the contrast of the scan. The gadolinium ions then accumulate in the kidneys for elimination. However, in cases of poor renal function, this can prove to be toxic. Patients with advanced chronic kidney disease, if injected with a typical gadolinium-based contrast agent, are at a significant risk of developing painful, debilitating disorders including contrast-induced nephropathy.

Just last year, a team at Massachusetts General Hospital devised a solution by proposing an alternative, safer manganese-based contrast agent (Mn-PyC3A) which not only produces identical contrast enhancement but is also made of an essential element which poses no risk to human health.

Hyperpolarisation
Within the past decades, although MRI has become indispensable in medical diagnostics, its sensitivity leaves much to be desired. In a typical MRI scan, due to the way that the molecular energies are naturally distributed, the MR signal from only 1 in 200,000 molecules is actually detected. Consequently, vital molecules (including potential cancer biomarkers) remain undetected due to their inherently low concentrations (<1mM) – severely restricting the scope of MRI.

Hyperpolarisation can overcome this sensitivity challenge by skewing the distribution of molecular energies and enabling detection of significantly more molecules. In hyperpolarisation, parahydrogen (polarised H2) or carbon-13 is chemically incorporated into an array of molecules (including glucose, urea, and pyruvate) and injected into the patient as tracers. This can increase the detected MR signal by up to 100,000 fold.

Hyperpolarisation drastically improves the imaging of metabolic activity – which is essential for tumour detection and monitoring. Cancer cells perform rapid glycolysis to obtain energy to fuel growth and division – converting pyruvate into the byproduct lactate in the process. A significant biomarker of aggressive, rapidly-growing tumours is a high lactate concentration. To assess the metabolic activity of a tumour, the patient is injected with carbon-13 enriched pyruvate and the concentration of hyperpolarised lactate produced is measured.

Unlike all other scanning methods, this technique yields information about not only the anatomy of a tumour but also its biochemistry – proving to be invaluable in assessing the efficacy of cancer treatment. Remarkably, within 24 hours of starting a new treatment, hyperpolarised MRI scans can reveal if a tumour is responding.

The figure above shows a prostate tumour before and only 48 hours after starting treatment. The MRI scans in the first row show a slight decrease in tumour size whilst the graphs in the second row (which refer to the detected concentrations of lactate and pyruvate) show that the lactate concentration has plummeted – highlighting the success of the treatment.

 

This biochemical change is further illustrated in the lactate concentration maps (last row) generated from the hyperpolarised MRI scans.

With MRI hyperpolarisation, clinicians can swiftly identify the ideal treatment for each patient based on the unique metabolism of the tumour – representing a major milestone in advances toward personalised medicine.

---

---

The figures quoted in this essay have been meticulously researched and documented. Below is the bibliography for this section of my essay:

Berger, Abi. “How does it work?: Magnetic resonance imaging.” BMJ 324.7328 (2002): 35. PubMedCentral (PMC). Web. [Last Accessed: 21 Feb 2019]

NIBIB gov. “How Does an MRI Scan Work?” Youtube, 16 Apr 2013, https://www.youtube.com/watch?v=1CGzk-nV06g&feature=youtu.be [Last Accessed: 21 Feb 2019]

Devlin, Hannah. “Introduction to FMRI.” Nuffield Department of Clinical Neurosciences, University of Oxford, [Last Accessed: 22 Feb] https://www.ndcn.ox.ac.uk/divisions/fmrib/what-is-fmri/introduction-to-fmri 

“MRI contrast agent.” Wikipedia, Wikimedia Foundation Inc., [Last Accessed: 22 Feb 2019] https://en.wikipedia.org/wiki/MRI_contrast_agent

“Contrast Dye and the Kidneys.” A To Z Health Guide, National Kidney Foundation, https://www.kidney.org/atoz/content/Contrast-Dye-and-Kidneys [Last Accessed: 22 Feb 2019]

Gale, Eric, and Peter Caravan. “Gadolinium-Free Contrast Agents for Magnetic Resonance Imaging of the Central Nervous System.” ACS Chem. Neurosci. 9.3 (2018): 395-397. ACS Publications. Web. [Last Accessed: 22 Feb 2019]

Freeman, Tami. “Medical innovations: A safer MRI contrast agent.” physicsworld: Focus On Biomedical Physics. June 2018: 6. Print. [Last Accessed: 22 Feb 2019]

“What is hyperpolarisation?” Centre for Hyperpolarisation in Magnetic Resonance (CHyM), University of York, https://www.york.ac.uk/chym/hyperpolarisation/ [Last Accessed: 24 Feb 2019]

Miloushev, Vesselin, Kayvan Keshari, and Andrei Holodny. “Hyperpolarization MRI.” Top Magnetic Resonance Imaging 25.1 (2016): 31-37. PMC. Web. [Last Accessed: 03 Mar 2019]

Iali, Wissam, Peter J. Rayner, and Simon B. Duckett. “Using parahydrogen to hyperpolarize amines, amides, carboxylic acids, alcohols, phosphates, and carbonates.” Science Advances 4.1 (2018). American Association for the Advancement of Science. Web. [Last Accessed: 24 Feb 2019]

Ward, Christopher, et al. “Noninvasive Detection of Target Modulation following
Phosphatidylinositol 3-Kinase Inhibition Using Hyperpolarized 13C Magnetic Resonance Spectroscopy.” Cancer Research 70.4 (2010): 1296-1305. American Association for Cancer Research. Web. [Last Accessed: 04 Mar 2019]

“Super-Cool Imaging Technique Identifies Aggressive Tumors.” National Institute of Biomedical Imaging and Bioengineering, National Institutes of Health, https://www.nibib.nih.gov/news-events/newsroom/super-cool-imaging-technique identifies-agressive-tumors [Last Accessed: 03 Mar 2019]

“Cambridge extends world leading role for medical imaging with powerful new brain and body scanners.” Research – News, University of Cambridge, https://www.cam.ac.uk/research/news/cambridge-extends-world-leading-role-for medical-imaging-with-powerful-new-brain-and-body-scanners [Last Accessed: 03 Mar 2019]

Peirce, Andrea. “Hyperpolarized MRI: A New Tool to Assess Treatment Response within Days.” Memorial Sloan Kettering Cancer Centre, Sloan Kettering Institute,
https://www.mskcc.org/blog/hyperpolarized-mri-new-tool-assess-treatment-response within-days [Last Accessed: 03 Mar 2019]

Image Credits:

T-state: Liddington, R. “PDB ID 1HGA.” 1992. Originally published in the Journal of
Molecular Biology 228:551. Reproduced in Lehninger PRINCIPLES of BIOCHEMISTRY
(Seventh Edition). Print.

R-state: Silva, M. “PDB ID 1BBB.” 1992. Originally published in the Journal of
Biological Chemistry 267:17,248. Reproduced in Lehninger PRINCIPLES of BIOCHEMISTRY (Seventh Edition). Print.

Ward, Christopher, et al. “Figure 6. Effect of everolimus treatment on GS-2 tumour
xenografts” Cancer Research, American Association for Cancer Research, Web,
http://cancerres.aacrjournals.org/content/70/4/1296

 

Return to the Stamataki Lab

I’ve kicked off this summer with a trip to Birmingham to meet with Dr Zania Stamataki. During my prior work experience here, I improved my wet lab skills by learning essential techniques such as splitting cell lines, performing serial dilutions, and imaging plates with confocal microscopy.  This year, I chose to hone my computational biology skills to best prepare me for a future in research; Dr Scott Davies introduced me to bioimage informatics through training me to work with the ZEISS Efficient Navigation (ZEN) microscopy software.

The ZEN software can be used to process the plethora of data that a single imaged experiment can yield. I worked with data from an experiment in which Huh-7 hepatocytes (from an immortalised liver carcinoma cell line) are co-cultured with live T-cells and imaged using confocal microscopy.

---

I personally found this sophisticated mode of imaging fascinating so I did a little more research into the bioengineering responsible for the high resolution and vibrancy of the  cellular interactions captured by the confocal microscope. Unlike the humble optical microscopes of my school, which only need light and a pair of eyes to visualise cells, confocal microscopes require far more exotic elements: lasers and fluorescent dyes.

 

GFP

One of the most widely used fluorescent biomarkers is green fluorescent protein (GFP) which is used in this experiment to highlight the cell membranes. GFP is a very important tool in research as it enables us to look beyond the cell surface and visualise organelles and protein movements. GFP has a unique sequence of the amino acids serine, tyrosine, and glycine which can be triggered to undergo chemical transformations in the presence of UV light. This special amino acid sequence is buried deep within the molecule’s barrel structure of intertwined protein sheets.

 

When activated, a cyclisation reaction occurs in which glycine forms a chemical bond with serine. This new closed amino acid ring spontaneously dehydrates and oxygen surrounding the GFP molecule forms a new double bond with the tyrosine – creating the fluorescent chromophore. In this way, GFP automatically assembles its own chromophore which can be used by researchers as tracking dyes. For instacnce, through genetic engineering, the GFP protein can be incorporated into the genome of specific cells or be used to label proteins of interest.  It is especially perfect for studying live cells since the alternative small fluorescent molecules (ex. fluorescein isothiocyanate) are highly phototoxic and inflict more damage upon live cells during imaging.

In the microscope images I worked with, GFP had been used to label the hepatocyte cell membranes. The T-cells were stained a vibrant blue whilst a cytoplasmic dye had rendered the hepatocyte cytoplasm red. The smaller dark regions enclosed within  cytoplasm are the hepatocyte nuclei.

---

Over the course of the week, I became proficient in producing orthogonal videos, which showcase the full depth of the hepatocytes, from multiple micrographs in a time lapse. Whilst processing these videos, I witnessed countless interesting phenomena; with Dr Stamataki’s permission, I’d like to share a few of my favourites with you.

Binucleate Hepatocyte Division

When I first starting working with the experiment data, Zania pointed this event out to me and I was both riveted and shaken to my core; this cell defied everything textbooks had me believe about cell biology! In this orthogonal video, a binucleate cell successfully undergoes mitosis to produce not two but three daughter cells.

Binucleate cells can be caused by cytokinesis failure in which the cell membrane fails to form a cleavage furrow so two daughter nuclei are trapped in the same cell. Alternatively, during mitosis, sometimes multiple centrioles form making the cell multipolar- pulling  the chromosomes in several directions and producing several nuclei in a single cell. Judging by how the binucleate cell proceeds to divide into an odd number of daughter cells, its original two nuclei probably arose from the latter mechanism.

Binucleation in vivo is pretty rare and, in a healthy human liver, such cells are incapable of dividing again, instead forced to remain in interphase indefinitely. Binucleate cancer cells, however, have a far different fate; more than 95% can undergo mitosis with a higher rate of chromosomal disjunction error – producing daughter cells with even more mutations. Despite binucleation rarely arising in healthy tissue, this phenomena is less of a rarity in cancer cells due to one of their unique quirks. A prominent biomarker of cancer cells is their multipolar spindles which increase the risk of binucleation and other mitosis abnormalities – leading to an accumulation of mutations in subsequent daughter cells in the cancer tissue.

Lamellipodia Network Formation

Inside the body, liver cells are always in contact with their neighbours with junctions between their lateral faces. In a lab culture, however, liver cells are seen tightly adhered to the plastic that they are grown on, practically clinging to the plate for dear life! The in vitro hepatocytes achieve this through multiple membrane projections known as lamellipodia.

Lamellipodia are made of the cytoskeletal protein actin and, in the video above, can be seen stretching out to not only tether carefully to the confluent monolayer but to also engulf some extracellular material. I’m mesmerised by the rapid yet tentative movement of these hepatocytes’ extensive threadlike network.

Another Kind of ‘Nuclear Fission’

The earlier orthogonal video of two nuclei becoming three was very surprising. In contrast, the phenomena below is absolutely incredible.  Two neighbouring T cells apply a mechanical stress on the hepatocyte’s nuclear envelope great enough to break the nucleus into two ‘daughter’ nuclei without the hepatocyte undergoing mitosis! I find it truly fascinating that the nucleus is not as static an organelle as it is so often portrayed; instead, it is dynamic and capable of distorting and even bifurcating in response to changing cytoplasmic pressures. 

 

 

---

As always, I’ve really enjoyed my time at the Centre for Liver Research this year and I’m so grateful to Zania for allowing me to learn so much about the cell biology and immunology of the liver and participate in her current work. Her team is currently in the midst of publishing some amazing cutting-edge research on a cellular mechanism similar to entosis; as soon as it is available to the public, I’ll share it here and, hopefully, write a review article about this interesting hepatocyte process!

WordPress does not support video uploads for this blog so the orthogonal videos in this post were compressed into gifs, hence losing resolution and quality. If you would like to see the original videos, I’ve also uploaded them to my YouTube channel:

• Binucleate Hepatocyte – Abnormal Mitosis
• Extensive Hepatocyte Lamellipodia Network
•  Nuclear Split

Bibliography

Kim, Dong-Hwee et al. “Volume regulation and shape bifurcation in the cell nucleus.” Journal of cell science vol. 128,18 (2015): 3375-85. doi:10.1242/jcs.166330

Shi, Qinghua; Randall W. King (13 October 2005). “Chromosome nondisjunction yields tetraploid rather than aneuploid cells in human cell lines” (PDF). Nature. 437 (7061): 1038–42. doi:10.1038/nature03958. PMID 16222248

Yang, F.; Moss, L.G.; Phillips Jr., G.N. “STRUCTURE OF GREEN FLUORESCENT PROTEIN.” Nat Biotechnol vol. 14 1246-1251 (1996). doi: 10.1038/nbt1096-1246. PMID 9631087

 

 

 

Give Me A Break: Bioengineering Solutions to Spinal Fracture

I’m writing this post in a lab at Queen Mary University in the heart of London. Through the EDT Headstart scheme, I have been given the brilliant opportunity to immerse myself in the field of bioengineering (from the physiological level right down to the molecular level) surrounded by leading academics and university lecturers as part of a 4 day residential course.

Absorbing fascinating new information about molecular bioengineering has been a major highlight of the course. On the first day, I attended Dr Karin Hing’s lecture ‘Engineering Biocompatibility for Bone Regeneration’ which inspired my independent group research project on the use of synthetic bone grafts, autografts, and growth factors in spinal fusion.

Biocompatibility lies at the intersection between biology and material science and it refers to the interactions between an artificial material and a living system. With its abilities to bear loads and control homeostasis as well as self-regulate and repair, bone is a particularly interesting natural material – but what is it actually made of? Bone primarily consists of inorganic materials, such as calcium hydroxyapatite, and organic materials, such as the protein collagen, which make up a rigid extracellular matrix. Within this matrix, resides a range of specialised bone cells: osteoblasts, osteoclasts, and osteocytes.

Like ourselves, bone has a life cycle. Our bones are constantly undergoing remodelling, cycling between phases of formation and resorption. In formation, mesenchymal stem cells in the bone marrow differentiate into osteoblasts (bone-forming cells). These osteoblasts create the extracellular matrix (known as the osteoid) and mineralise it with calcium deposits. Next, they find themselves trapped within the bone matrix and they mature into osteocytes which are no longer capable of bone formation, instead serving as a communication port- relaying signals between other bone cells (much like neurons).

Life is centred on balance, a yin and yang of sorts; as new bone is built up, old bone is destroyed through the process of resorption. Osteoclasts are a key player in resorption and prevent new bone formation from getting out of hand. Bone tissue contains rich deposits of minerals such as calcium and, when it is resorbed, the minerals leach out and enter the neighbouring blood stream. Through selective resorption of bone tissue, osteoclasts maintain homeostasis and regulate our calcium levels. Between formation and resorption is the reversal phase in which mononuclear cells inhibit osteoclasts and prepare the bone surface for osteoblasts to build the connective tissue up again.

With its constant repair and remodelling, bone can automatically fix virtually any hairline fracture or microdamage. But when it comes to breaks larger than 2.5 cm (known as critically large defects), the bone can no longer spontaneously heal and will require medical intervention. For instance, in cases of vertebrae fracture, spinal fusion is necessary to heal multiple vertebrae into a single, solid bone – restoring stability to the spine.

What do we use to bridge the gap between vertebrae? That’s where bone grafts come in. Traditionally, a bone graft involves removing a small portion of bone from the hip (where it will naturally heal back) and transplanting it between the vertebrae to promote bone growth at the fusion site. If the graft comes from the patient’s own body (which is currently the gold standard in spinal fusion) then it is referred to as an autograft. Allografts come from a cadaver or live donor and are used when the fracture is too large to bridge with the patient’s own bone.

harvesting an autograft from the iliac crest (hip)

The reason autografts are considered the gold standard is because they can integrate into the patient’s fracture more rapidly and completely and there is no risk of immune rejection or bio-incompatibility. However, autografts require the patient to undergo two surgeries – giving rise to an array of new complications including donor site morbidity and secondary infections. Additionally, multiple incision sites may be necessary to extract an ideal autograft which increases the risk of nerve injury and bone damage at the donor site. But allografts are not without their flaws either; they come with a risk of immune rejection and disease transmission and there are long waiting times to even obtain an allograft due to a shortage of eligible donors.

Recently, synthetic alternatives to bone grafts (such as calcium sulfate and collagen) have emerged. Initially, these synthetic bone grafts were bio-compatible but still did not promote bone healing. Let’s take a closer look into what is necessary for a graft to support successful bone formation. Firstly, we have osteoconduction which is the graft’s ability to support the attachment, migration, and growth of osteoblasts within the structure. Then, there is osteoinduction which refers to the graft’s ability to stimulate stem cell differentiation into bone cells (osteogenesis).

Ever tried gardening? The properties of an ideal bone graft are similar to those needed to grow a plant. Osteoconductivity (the scaffold properties of the graft) can be seen as the pot and soil, osteogenesis by the actual seeds, and osteoinductivity is best represented by the fertilisers and water that stimulate plant growth.

At the very least, a synthetic graft must be osteoconductive for a successful fusion. On the third day of the course, I had the chance to enhance my wet lab skills via testing protein adsorption in a range of synthetic grafts. The experiment involved incubating five types of graft material (both organic and inorganic) in an albumin solution and using a colorimeter and calibration curves to assess the efficiency of protein adsorption in each of the graft types.

The first synthetic grafts were bio-compatible but also bio-inert meaning that they had were osteoconductive but not osteoinductive. To overcome this issue, subsequent designs of synthetic grafts tend to be infused in growth factors. Growth factors stimulate differentiation of stem cells and growth of living tissue – accelerating bone formation.

The most widely studied growth factor is bone morphogenetic protein 2 (BMP) which acts as a chemotactic agent by recruiting stem cells to the fusion site and prooting their differentiation. BMP even stimulates the growth of blood vessels (angiogenesis) through the damaged bone which helps support new bone formation.

the structure of BMP-2

BMP is considered safe for use in spinal fusion and eliminates the need to take a large portion of the patient’s own bone cells in a graft. The only downside is that it is currently very expensive – increasing spinal fusion surgery costs by up to $15,000.

BMP works alongside, and can be incorporated into, both types of graft. Synthetic grafts can be made of bioabsorbable polymers (such as lab-grown collagen or calcium sulfate structures) which carry doses of growth factors. As the material dissolves in the body, the BMPs are gradually released, helping sustain bone growth. In autografts, the patient’s own cells can even be genetically engineered to express specific growth factors in greater quantities.

It has become increasingly clear that bone grafts cannot be developed solely from an engineering perspective. As Dr Alvaro Mata explained in his lecture on bioinspired materials, nature is a particularly important source of inspiration for designing any bioengineered system, particularly synthetic bone grafts. Structure at even the smallest levels, such as the arrangement of structural protein nanofibres, affects cell signalling and growth to a large degree.

Biomimetic design is developed through an appreciation of nature’s own design. Through dissecting a bovine metacarpal joint, I gained a far better understanding of the interplay between bone and cartilage that makes smooth movement effortless.

---

As the end of my week at Queen Mary’s bioengineering department crept closer, my team began to prepare a presentation showcasing our independent research in the fascinating field of spinal fusion. This presentation was the perfect culmination of the amazing molecular bioengineering techniques I had studied and carried out in the lab. To my absolute delight, we were awarded the 1st place prize for our level of detailed research and our debate style of presentation – a rewarding end to an incredible week!

Part II: Ultrasound Technology

In sharp contrast to the potentially mutagenic radiation emitted by x-ray based scanning devices, ultrasound has a long-standing reputation of being extremely safe – to the extent that it has even been trusted with imaging the delicate structures of the developing foetus. The surprising truth about this medical scanning modality so commonly associated with growth and new life, is that it was developed in response to a tragic mass death.

In the aftermath of the sinking of the Titanic, physicist Paul Langevin invented the world’s first transducer, the hydrophone – a scanning device which could detect hidden threats in the depths of the sea. Inspired by Langevin’s innovation, in 1942, neurologist Karl Dussik designed an ultrasound transducer capable of detecting hidden threats of tumours lurking in the depths of the brain. From that moment onwards, the field of medical ultrasound scanning blossomed.

The 1970s saw major advances in ultrasound technology which made it possible to view the organic structures of the human body as they truly are – dynamic, changing forms that can only be fully understood through their movement.

Doppler Ultrasound
The Doppler ultrasound was one of the first medical scanning devices designed to monitor the flow and velocity of blood. The transducer emits a series of pulses at the blood vessel; whilst the echoes received from stationary tissue are identical from pulse to pulse, the time taken for the echoes to return from the blood vary slightly with each ultrasound pulse. The resulting Doppler frequency is observed and can be used to calculate the blood velocity with the following equation

Blood velocity abnormalities can indicate the presence of blockages, including blood clots and vascular tumours, radically improving diagnosis of a range of conditions from deep vein thrombosis to atherosclerosis.

Ultrasound Needle
Most recent developments in ultrasound technology have been focused on designing portable devices perfectly suited to niche, yet critical, situations. An all-optical ultrasound transducer, which can be fitted into the tip of a needle, has recently proven successful in a preclinical swine study by researchers at University College London and Queen Mary University of London.

Currently, during keyhole procedures, surgeons are forced to rely on external ultrasound probes and pre-operative scans to visualise soft tissues and structures. This minimally invasive ultrasound device images tissue in real time to assist in surgery.

The compactness of the ultrasound transducer is crucial for its use in keyhole surgery. Instead of a bulky electronic transducer, the ultrasound needle uses an all-optical transducer which exploits the photoacoustic effect (in which ultrasound waves are formed when light is absorbed). The tip of the needle consists of two 300μm optical fibres. The transducer optical fibre is coated in a multiwalled carbon nanotube polydimethylsiloxane composite capable of absorbing laser light and generating ultrasound pulses. These pulses are reflected off the tissue and detected by the receiver optical fibre which converts the ultrasound echoes back into light.

---

---

The figures quoted in this essay have been meticulously researched and documented. Below is the bibliography for this section of my essay:

Constantin, Chilowsky, and Paul Langevin. Production of submarine signals and the location of submarine objects. US 1471547 A, United States Patent and Trademark Office, 19 May 1917. [Last Accessed: 05 Feb 2019]

Deane, Collin. “Doppler Ultrasound: principles and practice.” Doppler in Obstetrics, ISUOG Educational Committee, [Last Accessed: 16 Feb 2019] https://sonoworld.com/client/fetus/html/doppler/capitulos-html/chapter_01.htm

Freeman, Tami. “Medical innovations: All-in-one imaging.” physicsworld: Focus On
Biomedical Physics. June 2018: 6. Print. [Last Accessed: 18 Feb 2019]

Finlay, Malcom et al. “Through-needle all-optical ultrasound imaging in vivo: a preclinical swine study.” Light: Science & Applications (2017) Nature. Web. [Last Accessed: 19 Feb 2019]

Image Credits:

Deane, Colin. Diagram of Doppler ultrasound. Doppler in Obstetrics, ISUOG
Educational Committee, 2002. Web.
https://sonoworld.com/client/fetus/html/doppler/capitulos-html/chapter_01.htm#fig02

Finlay, Malcom et al. Schematic and inset photo of sharp inner needle. Light: Science
& Applications, Nature, 01 December 2017. Web. nature.com/articles/lsa2017103.pdf

Scanning Devices for Medical Applications – Part I

This is the essay commended by judges from Newnham College, Cambridge. I am delighted to share with you this work that took me almost a month to write and far longer to research. I loved the experience of writing this essay and immersing myself in this fascinating topic.

However, it is important to note that, since this was a 2500 word academic essay, it does make for heavier reading than my previous articles written specifically for this blog.

So brew some tea, find a comfortable reading nook, and settle in for this comprehensive exploration of the field of medical scanning from its origins over a century ago to its potential for the coming decades.

---

PART I : ORIGINS AND X-RAYS

---

The disciplines of engineering and medicine have long been intertwined, with physics concepts supporting the development of diagnostic tools and clinical demands fuelling exciting innovations in engineering.

Scanning devices for medical applications, all of which consist of an emitter and sensor, lie at the very heart of the intersection of these disciplines. Since its inception, over a century ago, this technology has been constantly evolving with new developments.

Origins
In a darkened laboratory in Würzburg, Germany (1895), an eerie glow emanating from a screen behind his cathode ray tube propelled Wilhelm Röntgen to discover invisible rays. These rays were capable of not only penetrating the heavy black paper insulating the Crookes tube but even Röntgen’s own skin. Astonished by his discovery of “a new kind of ray”, Röntgen began to experiment with the penetrating abilities of the x-ray to produce one of the first radiographs: his wife’s hand adorned with her engagement ring.

Shocked by the sight of the internal structures of her own hand, she reportedly gasped, “I have seen my death”. Despite such a macabre initial critique, x-rays captivated the world’s attention and optimism for their applications; Röntgen’s moment of serendipity gave rise to the ceaselessly evolving field of medical imaging.

X-Ray Modality Technology
X-rays are passed through the patient and are absorbed, to varying degrees, by different tissues in their body. In indirect flat-panel detection, unabsorbed x-ray photons are transmitted through the body and strike caesium iodide crystals in the first layer of the detector. The crystals scintillate when excited by radiation – emitting flashes of visible light proportional to the energy of the x-ray photons. Amorphous silicon photodiodes in the detector’s second layer transform light into electrical charges to produce digital images.

X-Rays and Cancer 
X-rays play an essential role in mammography; pre-cancerous breast tissue contains clusters of microcalcifications (minor calcium hydroxyapatite deposits) which act as major indicators of accelerated cell division. Microcalcifications can go undetected in biopsies. However, in x-ray scans, these deposits are clearly revealed as bright white specks due to their very high x-ray absorption – enabling early detection.

CT Scan
The humble x-ray also forms the basis of the revolutionary multisection CT scanner (a scanning device introduced in 1992 that aims a narrow beam of x-rays at the patient in a complete spectrum of angles); this device outputs multitudes of cross-sectional ‘slices’ (tomographic images) which can be digitally ‘stacked’ to produce three-dimensional models. The technique improves both diagnostic accuracy and spatial resolution in the Z axis due to the creation of thinner tomographic slices. This also reduces the risk of partial volume artefact formation (imaging errors which occur when tissues with varying thickness and radiation absorption are in close proximity).

EOS Scan
The latest x-ray modality technology addresses a major drawback of CT scanning: its significant radiation exposure. The EOS scanner, developed in 2007, is a vertical scanner consisting of two pairs of perpendicular radiation sources and detectors. This allows for precise 3D reconstruction as well as imaging of the skeletal system in its natural weight-bearing posture. This innovation helps clinicians identify and diagnose subtler issues which are only apparent in the patient’s skeletal system in the context of its typical load-bearing alignment (as opposed to its relaxed alignment when the patient is lying down in a CT scanner).

The implementation of revolutionary particle physics technology allows high-resolution (254 μm) radiographs to be taken with significantly less radiation exposure than CT scanners. The EOS scanner uses a proportional multiwire chamber (first developed by the Nobel Prize winner Georges Charpak) which enables extremely sensitive detection of even single x-ray photons – allowing a significant reduction in x-ray exposure without compromising the resolution.

This unique aspect makes it particularly suited for monitoring scoliosis in adolescents. Since scoliosis is eight times more prevalent and more severe in females than in males, adolescent girls are typically exposed to excessive radiation from the multiple CT scans (up to 5 per year) necessary to assess the progression of this condition. High level of radiation exposure, especially during a critical phase of rapid growth, increases the risk of developing cancer; growing breast tissue is highly sensitive to the carcinogenic effects of ionising radiation which cause a 69% increase in breast cancer incidence.

The EOS scanner, with its significantly lower radiation exposure, is proving to be vital in limiting the risk of cancers whilst safely monitoring scoliosis.

 

---

The figures quoted in this essay have been meticulously researched and documented. Below is the bibliography for this section of my essay:

“History of Radiography.” NDT Resource Centre, National Science Foundation,
https://www.ndeed.org/EducationResources/CommunityCollege/Radiography/Introduction/history.htm [Last Accessed: 28 Jan 2019]

“X-Ray.” Wikipedia, Wikimedia Foundation Inc., https://en.wikipedia.org/wiki/X-ray [LastAccessed: 28 Jan 2019]

Cruz, Robert. “Digital radiography, image archiving and image display: Practical tips.” Canadian Veterinary Journal, 49.11 (2008): 1122-1123. PMC. Web. [Last Accessed: 04 Mar 2019]

Wang, Zhentian, et al. “Non-invasive classification of microcalcifications with phase-contrast X-ray mammography.” Nature Communications 5 (2014). Nature. Web. [Last Accessed: 01 Mar 2019]

Rydberg, Jonas et al. “Multisection CT: Scanning Techniques and Clinical Applications.” RadioGraphics 20.6 (2000): 1787-1806. RSNA. Web. [Last Accessed: 30 Jan 2019]

Bell, Daniel, and Vikas Garg et al. “Partial volume averaging (CT artifact)” Radiopedia, Radiopedia, https://radiopaedia.org/articles/partial-volume-averaging-ct artifact-1?lang=gb [Last Accessed: 30 Jan 2019]

Illés, Tamás, and Szabolcs Somoskeöy. “The EOS (TM) imaging system and its uses
in daily orthopaedic practice.” International Orthopaedics 36.7 (2012): 1325-1331.
ResearchGate. Web. [Last Accessed: 31 Jan 2019]

“Information and Support.” NSF, National Scoliosis Foundation,
https://www.scoliosis.org/info.php [Last Accessed: 02 Feb 2019]

Morin-Doody, Michele et al. “Breast cancer mortality after diagnostic radiography: findings from the U.S. Scoliosis Cohort Study.” Spine 25.16 (2000): 2052-2063. NCBI. Web. [Last Accessed: 02 Feb 2019]

Image Credits:

Röntgen, Wilhelm. “Hand mit Ringen (Hand with Rings).” Wikipedia, Wikimedia
Foundation Inc., 22 December 1895.

Hendriks, Eva, et al. “Regression of Breast Artery Calcification.” Cardiovascular
Imaging, Journal of the American College of Cardiology, August 2015.
http://imaging.onlinejacc.org/content/8/8/984/F1

EOS Imaging, Paris, France. A reconstructed 3D model of the full vertebrae of a
scoliosis patient based on an EOS TM 2D examination. Reproduced by Illés, Tamás in
International Orthopaedics 36.7 (2012): 1325-1331. Web. https://www.researchgate.net/figure/Full-body-surface-reconstructed-3D-model-based-on-an-EOS-TM-2D-examination_fig3_221867288

 

Visual Prosthesis : Latest Innovations

The face of biotechnology is constantly evolving; blink and you’ll miss it. Latest developments in visual prosthesis are bringing us one step closer to the miracle of restoring sight to the blind.

Over a year ago, I blogged about one of my many visits to the Wellcome Museum in London and, in particular, my interest in the glass eye collection of W. Halford (1870-1920). Curiosity had compelled me to further research the history of ocular prosthetic devices. One of the earliest ocular prostheses dates back to 2900–2800 BCE and serves a purely aesthetic purpose with its radiant “gold lines patterned like sun rays”.

Modern ocular prostheses have become safer and more lifelike. They are typically made of medical grade plastic acrylic in order to avoid issues such as shattering of the prosthetic in the eye. Realism of artificial eyes have been continually developing since the first ocular prosthesis; the main limitation being realistic pupil movement. Ocularists and eye surgeons have always worked together to make artificial eyes look more realistic. For decades, all efforts and investments to improve the appearance of artificial eyes have been dampened by the immobility of the pupil. The most recent solution to this issue is to use an LCD display to vary the pupil size depending on the light intensity of the environment. Despite these large strides forward in the aesthetic qualities and features of ocular prosthetic devices, until very recently, such devices have not offered the most valuable feature of all: vision.

Last night, in the most recent issue of the National Geographic which focuses on the future of medicine, one article in particular caught my eye; researchers have made the possibility of a true bionic eye tangible through 3D printing.

The concept of electrically stimulating the visual cortex and retina to produce artificial sight first surfaced in the 18th century, however (due to technological and anatomical difficulties) this concept is only reaching actualisation within the past decade. The retina  contains light detecting cells called photoreceptors which convert light energy  stimuli into electrical energy for transmission to the brain. In most cases, blindness is caused by the death of these photoreceptors as opposed to the transmission nerve cells that respond to stimuli and send messages to the brain. While those nerve cells cannot detect light on their own, they can respond to electrical stimulation; this is what much of the research into visual prosthesis is focused on.

The Argus II Retinal Prosthesis (patented in 2011) is the first of such devices to receive FDA approval; it consists of a camera mounted onto eyeglass frames which is connected to a processor and retinal impant containing 60 miniscule electrodes. Although certainly a blessing to patients who have been able to regain sight with this prosthesis, the Argus II is far from a natural bionic eye.

              

 

The  latest reported prototype of the bionic eye (developed by researchers from the University of Minnesota in August 2018) has an uncanny realism and functionality that I found entirely fascinating. The bionic eye has a 25% efficiency in converting light into electrical signals thanks to its semiconducting polymer photodiodes which cover a 3D printed base of silver particles lining the interior of the hemispherical glass dome. This prototype represents a milestone in the field; it proves that semiconducting materials can be ‘printed’ onto curved surfaces and, more importantly, that such devices can be cost-effective- designed by a humble 3D printer in under a hour as opposed to in high-tech microfabrication labs.

Beautiful as it is, this bionic eye still needs further development to make the surface softer and more suitable for implantation and to improve light conversion efficiency. Within another decade, such devices should be available as true cures to blindness- perhaps the future isn’t so dark after all…

 

What’s in a baby’s kick?

I am delighted to have been able to attend a Royal Institution ‘Summer School’ workshop on bioengineering this August. The focus in this workshop was on refining drug delivery systems and using mathematics to determine whether a foetus is developing properly.

---

The first half of the session was lead by Rachel Dorris, a medical physicist, and was centred on how drugs can be better designed for use in inhalers.

Asthma is an autoimmune disease caused by the spontaneous contracting of the smooth muscles that surround the bronchioles in the lungs which results in constricted airways. Inhalers contain a drug (a bronchodilator) that can ‘open up’ the airways.

 

(credit: slideplayer)

In order for this treatment to be effective, the drug uptake (the retention of the drug in the necessary organ) must be at its optimum. To measure the drug uptake, a medical imaging technique that suits this purpose must be chosen. After some collaboration, our team decided that a Nuclear Medicine Scan was the ideal technique. It involves giving the patient the bronchodilator with a small amount of a radioactive tracer (Technetium-99m) and then observing the patient with a gamma camera. The image quality is relatively low but provides functional information and can be easily used to observe drug uptake.

This is an example of a Nuclear Medicine Scan of the lungs. The whitest areas show the highest drug uptake. (credit: slideshare)

From Nuclear Medicine Scans of 4 different drug particle sizes in patients’ lungs, we performed a qualitative analysis of the images- determining which particle size we believed would have the optimum drug uptake in the bronchioles. We then performed a quantitative deposition analysis of the images to confirm our ideas. The conclusion we arrived at is that a bronchodilator with a particle size of 1.5μm was best suited to use in inhalers.

---

The second half of the workshop was lead by Dr. Stefaan Verbruggen, who is currently researching the effects of kicking in the womb during pregnancy on the development of musculoskeletal diseases, and, in particular, developmental dysplasia of the hip (DDH).

The hip is a type of “ball-and-socket” joint. In a normal hip, the ball at the upper end of the femur fits firmly into its socket in the pelvis. In babies with DDH, this joint is not properly formed and so the ball does not fit well in the socket and is easily dislocated. In all cases of DDH, the socket (acetabulum) is shallow which means that the ball cannot fit inside it comfortably.

 

(Credit: Oxford University Hospitals)

1 in every 1000 babies are born with a form of DDH. There are varying degrees of severity in DDH cases:

• Subluxatable- the ball of joint is simply loose in the socket
• Dislocatable- the ball lies within the socket but can be dislocated easily
• Dislocated- the ball is completely out of the socket.

(Credit: helpmegrowutah.blogspot.co.uk)

DDH is not genetic, but the risk of a baby having DDH greatly increases if there was not enough room in the womb whilst its bones and joints were developing.

A foetus’ bones and muscles, like in adults, react to stress by growing. In adults and children, the main stressor acting on the hip joint is gravity, but this is not the case in foetuses. Instead, the main way that foetuses can ‘exercise’ their hip joints, is by kicking. Considered one of the most endearing interactions between a foetus and the outside world, kicking is vital to its development; if there is not enough room for a foetus to kick, complications with joint development and DDH can arise.

• It is a ‘breech baby’ (the foetus is positioned feet first). In scans, it is revealed that these babies cannot fully extend their legs when kicking.
• There are issues with oligohydramnios (in which there is less amniotic fluid- normally due to a fault in the foetus’ urine production so that it ingests the amniotic fluid but does not pass it out). Notice how, in scans, these babies don’t appear to kick at all.

Cine-MRI scans of baby kicks

 

Curiously enough, foetal twins tend to have enough space to kick comfortably, despite having to share the womb. (They also are unable to kick each other due to a membrane wall separating them!)

---

The progress of a foetus’ joint growth and muscle activity is measured to ensure that the pregnancy is going well and to calculate the risk of DDH. However, when foetuses are in the womb, experimental equipment, naturally, cannot be used and only images can be used. This means that mathematics must be employed in a fascinating manner to calculate strength of a kick from only two MRI scans.

 

Start of kick (Frame 76)

End of kick (Frame 83)

First we must work out the change in angle from the start to the finish of the kick. At the start of the kick, the angle between the hip and the ankle is 87°. At the finish, it is 153°. This means that the change in angle is 66°. Then, this must be converted to radians, making the value 1.15192 radians.

The MRI scan was taken at 3 frames per second and the kick took 8 frames. Using this, we can calculate that the length of the kick in seconds is (to three decimal places) 2.667 seconds.

Speed is found by dividing the distance by time taken to complete the distance. In this case, the distance is the change in angle. Therefore, the speed (or velocity) of the baby’s kick is  1.15192 / 2.667 = 0.432 radians/s.

The length of the lower leg is 58.4 millimetres and it can be considered a radius in this case. Acceleration is found by multiplying the square of the velocity by the radius. So the acceleration of the kick was 0.432² x 58.4 = 10.899 mm/s²

The foetal leg would weigh 0.321 kg. Since force = mass x acceleration, we can now also find the force that the foetus must generate to perform this kick- 3.498 N.

This example demonstrates just how powerful mathematics is; using only 6 relatively simple calculations, we can quickly work out the force of a foetus’ kick from just 2 basic MRI scan frames.

---

Overall, the workshop was an incredible introduction into the field of bioengineering. Without a doubt, the second half of the workshop was my favourite as I found the exploration of DDH in babies very interesting and I was amazed by how easy a seemingly impossible task was made by using formulae that I had already learnt in school but had never seen being used to solve ‘real-life’ problems.