Elementary, my dear Watson

One of the enduring mysteries in Arthur Conan Doyle’s delightful Holme’s series is that whether Watson was needed at all. However, it transpires in the text that Watson is more like a second person to whom Holmes is speaking with (much like a sounding board).

Purists although would point out the truth hiding in plain sight.

In real life, the modern day Watson is far from perfect, and its cheerleaders sound even dumber.

In a scathing write up in Wall Street Journal, the reporters have laid bare the hype behind IBM Watson that was supposed to have revolutionised cancer treatment. Doctors from a reputed hospital joined the project to “provide direction” followed by a series of photo-ops. The word out was that we were, at last, on the brink of revolutionising the healthcare as we knew it.

To cut a long story short, it didn’t happen. It was being used for marketing by prominent hospital chains with “opinions” sought from the “latest literature”. There is NO scientific evidence that it made an iota of difference.

I have no idea what is the problem that they were trying to solve in the first place. Everything is available off the shelf, and it is incredibly easy to start off a chemotherapy regime. The promised efficiencies of the scale never materialised. Watson was supposed to tap into existing electronic medical records so that it can pre-fill and auto-populate the fields. However, these are limited to restricted geographies and hence do not have worldwide relevance. They do not have a mechanism to quantify the outcomes and measure responses. How would they know what works and what doesn’t? If the results have to be determined by the clinician, what is the point of having this expensive system in the first place? Another issue with it is its poor adaption to clinical ontology. I have seen and heard about the way it struggles with building up the “clinical dictionary” that makes it difficult for it to adapt to hand-written notes. It is likely to lose out all the context of patient’s clinical history. How accurate would its recommendation be if there are significant gaps in the patient’s verbal history? Clinicians often make educated guesses and implied understanding of patient’s issues.

I remain sceptical about the broad-based “research” that multinational corporations push out. There seems to be no truth in the advertising. The original biomedical research has plateaued concerning outcomes. If we have run out of targets, we dig deeper to find new ones! How much do we need to “understand” about pathogenesis? Let us find something worthwhile from what we have discovered. We require an audit of existing “research” to prevent overlaps.

Instead, we need similar and parallel approaches in patient support and ancillary services. The use of mobile technology with adequate privacy safeguards to assist affected patients is promising.

Even from the perspective of Radiation Oncology, we need more to understand this fascinating science. Radiation Biology hardly ever scores funding; for example, we need a scoring system to quantify plans and qualitative analysis for clinical outcomes. We also lack standards to match it with reported clinical results.

The undue focus on the biomedical research will rob us away from our real targets- the patients. In the quest for mutations, we forget its the real people who need good palliative care and support system.

Imagine, if we could channelise the millions of dollars for something constructive! I strongly feel that it is a wasted effort. AI in healthcare can wait.

Achieving the digital detox.

Over the past few months, during my interaction with various people on Twitter and offline, I have been hearing about the information deluge that makes it impossible for them to acquire new skills. We indeed have limited 24 hours!

I wouldn’t be able to give a blow by blow account of how I manage things, but I have had to stick to certain good habits that have made things more comfortable for me. I will mention a few services below that make it incredibly easy to flow past the flood of distractions.

1) Mobile phone:

This is the biggest annoyance! I am off the social networks on the device barring Telegram. More on that later.

Twitter is accessed only on the browser. No dedicated applications exist. All notifications on the device are blocked except for text messages.

Telegram helps me to mute all conversations except people whom I deem essential. I have a lot of channels where I consume content passively. No mainstream social networks like Facebook or Instagram for me. They are antiquated because I cannot control them.

I also don’t have a fear of missing out. My associates either call me or text me if needed. Android has become better to manage notifications in recent times. I don’t have any experience with iOS, but I remain convinced that Apple iPhones are merely iPods with a calling facility.

2) Email.

I had mentioned this earlier too. I use Fastmail because I find there is inherent value in paying up for the email. I use a lot of aliases whenever I sign up for the service. It helps to signup with a unique email address. For example, for Dropbox, my alias will be dropbox (at) FastMail dot com (it is a hypothetical- just for illustration). Therefore, if any spam flows into my inbox, I know where is the leak from. All I have to do is to delete the alias.

This simple hack has served me well over six years, and I am happy to stay with this service. Mainstream email applications like Gmail or Yahoo are useless.

I have also created extensive rules which directs the email in the trash. It helps to clean up the clutter at the server itself without manual intervention. For example, all newsletters go to trash directly. Some of them are automatically marked as read and stay in the inbox- I scan through them when I get time. When they are marked as read, I don’t get a notification. Therefore, can easily stay focused on my work without being distracted by the flow in the inbox.

3) Password Manager

1Password is the password manager that is my life saver. It generates unique passwords for all the websites. It is a paid service, but good cloud sync helps me to sync it with my Android device as well. It eliminates the need to remember unique passwords.

4) The use of Telegram chat app

Telegram remains the only way to stay connected with any semblance of “social network”. I use a combination of groups and channels to stay informed. Channels work as public broadcasts. Any specific information I need is transmitted to it. I use bots (both paid and free) to achieve the effect.

For example, I use the IFTTT bot to work with the RSS feeds to populate the channels with the Pubmed content. If I need to track, say the latest publications in the development of MR-LINAC, I don’t have to visit the website manually. By use of booleans, I can filter the content, generate the specific RSS feeds which pipes it elegantly in the channel via IFTTT bot. Likewise, I use junction-connection bot and Feed-Reader bot for different purposes. I pool in information from all specific channels I need to follow into one omnibus channel so that I don’t have to deal with a multitude of channels. I do this by using junction-bot on Telegram.

Feed-Reader bot helps me to tap into various other social networks. For example, I have a specific channel devoted to cycling. All posts from multiple Instagram accounts flow in the channel. It helps me to keep track of the sectoral development. Likewise, I developed a channel for journalists on Telegram to keep track of telecom sector and clean energy. I also have a dedicated art channel that I helped to make for a friend. That collects all impressionist art, beautiful nature pictures and graffiti! None of the posts is done manually.

Focused groups require extensive group management. I recommend using Combot because it comes with a beautiful web-interface. Although the community bot management has introduced a paid plan, it is free for groups that have up to 100 members. The bot deletes specific stop words automatically along with other nifty features like muting users. The bot also keeps groups free of spam messages. Therefore, the groups stay efficient, productive and on course. It is unlike WhatsApp where users start spamming others without any rhyme or reason.

This above may sound onerous, but it helps to maximise the efficiency gains. As long as you are not distracted, it helps to keep focused on work.

In the busy schedules that we keep, always find time for solitude. That is the most critical period to stop and reflect on your goals.

Digital tools need a constant refinement. Hopefully, I will update this in the future.

(Images are for representational purpose only. This blog post is not intended for any commercial purpose).

Brain Tumours Bot for feedback

I am happy to announce the launch of a feedback bot to collect link to various brain tumour charities across the world (braintumourbot).

I had toyed with the idea to create a Wiki, but I realised that existing tools are too messy that can be utilised effectively. Why not have something that makes things more efficient?

It is how you do it:

1) Install Telegram and open it.

2) Search for @braintumourbot

3) Press “start” at the bottom. Then type it in the space provided.

4) In the bot description, there exists a link to the channel.

5) Your submissions will be updated there.

You can search for respective charities either in the search box in the channel or typing particular hashtag. For example, in the link provided, the hashtag for Britain is #UK, and the city hashtag for London is #LON.

The Brain Tumour support channel is also active! (@cnssm)

I will provide a complete name for cities that are far away from the urban centres.

This bot on Telegram chat application is the first ever crowdsourced experiment to get everyone on board.

Wiki for brain tumour charities

Early this week, I had announced that I would develop a Wiki page for collecting the links to all brain tumour charities. However, this is a herculean task indeed.

If I were to use the existing resources at Wikipedia, editing the webpage is an onerous task. They haven’t migrated to a simple modular structure like in a blogging template nor do they allow any specific linkages with applications on the desktop. The only way out is to use a browser. Their interface would scare the most battle-hardened veteran as well.

What is the way out? Telegram again comes to my rescue. The way out is straightforward.

1) I have created a submission bot (braintumourbot) that will interface directly with a private group.

2) A standard format for submission can be evolved.

3) I will be using specific hashtags. For example #US for North America, #UK for Britain, #AUS for Australia and the likes. Any user keen to look for information under the head will be able to locate it by just clicking on the hashtag.

4) #LON under London will display the charities based there. If they a website address and a Twitter/Social account, it will be listed accordingly. This consistent storage of information will help me to organise the information quickly.

Telegram does offer the speed and reliability with added security for the users. I think this is the best way forward to crowdsource the information.

I have ideas for a dedicated mobile application as well, but that would require yet another mobile app; resources can be utilised to make things efficient. I remain sceptical about either a website or the apps.

Let’s see how this experiment grows!

Biomedical research, outcomes and survivorship

This post was the result of being tagged in an interesting discussion on Twitter about survivorship. I want to give in my viewpoint.

Survivorship isn’t just about a hard clinical outcome. It is about the functional autonomy of an individual. There are various issues tied up here; it wouldn’t be appropriate to discuss survivorship issues in isolation. I would, of course, take the example of brain tumours per se and weave in the narrative for various interconnected problems. I have been thinking about it for the past two weeks and here’s my observation.

I was lucky to attend Paediatric Neurooncology conference- which was my first trip to States. It had been a fantastic experience; great organisation as well as well attendees from different parts of the world. However, one thing stood out- the predominance of biomedical research. I wouldn’t hesitate to call it as vanity research because it is an arms race to practically nowhere. We have expended billions of dollars in hair-splitting pathways and identifying newer targets, but as an opportunity cost (investment versus tangible outcomes) it is a downward spiral and a wasted opportunity. I am not advocating less of research but one that has measurable practical results.

Let’s look at survivorship from three examples. Meningiomas, Medulloblastomas and DIPG (as the extreme). Meningiomas are mostly indolent but are considered “curable” after definitive surgery followed by an indication for radiation therapy. How and why the need for drugs has come in? What proportion of patients needs it? Are they effective? And if the disease is aggressive and progressive, when do we call it a day and suggest best supportive care for the patient? The hapless patient would need dependent care (based on what neurological functions have deteriorated). How do we define survivorship here? Have we made the patient functionally autonomous or helped the patient to adjust to the disabilities?

The “middle rung” (as I would call it because it has a decent prognosis) is medulloblastomas. Elegant research has divided this into various subtypes with the promise of “de-escalation” of treatment. How much, do we know that “less is less” and not “really less” of radiation therapy? Chemotherapy, from the classical radiobiological constructs, eliminates only the most active cell populations but still, the “spurters” remain in the cell pool. Reducing the radiation dose will, in all probability, really compromise with the eradication. The “myth” of radiation-induced side effects continues to this day without really accounting for the long-term neurocognitive effects of chemotherapy.

What is survivorship in this context?

The last but not the least is relative rare diffuse intrapontine glioma. This diagnosis is universally fatal, and despite an intense outpouring of research, the outcomes haven’t improved. They have drilled catheters to deliver drugs right at the source and have claimed “success”, but the overall scenario remains bleak. What is survivorship here for DIPG?

Hence, either way, you look at it, there are no easy answers. The spurt of biomedical research (often cornering the most significant resource) needs to be tempered with the realistic expectations of the pharmaceutical industry (that funnelled research into practical, actionable targets). There have been clamours, of course, for a “close collaboration” between the industry and academia but everyone is aware of the pot of gold. An actionable mutation followed by a drug and patent protection for about ten years is equal to profits. Insane profits. But, how has survivorship improved? Instead, we have newer metrics to measure “survival” like “progression-free intervals” which has no meaning because the disease is always present.

I feel that it is important to pay equal importance to the emerging role of technology and patient support. Like the innovative use of chat applications, the emergence of bots and various platforms that can make life easier to adjust with disabilities. Patient support is an ignored criterion that could get a better impetus and more funding to make lives more meaningful.

The central question remains- when to introduce “palliative care” and “hospice” in the evolution of the disease. These two questions determine the meaningful survivorship.

As from the preceding discussion, it is not easy to quantify survivorship. The goal of research should remain improvement in population outcomes. Cancer aetiology points out towards mostly preventable causes- air pollution, smoking etc. What are we doing to improve our score in that direction?

Last but not the least is cancer prevention. Sadly, it is not relevant for gliomas save for the fact that mobile phones are “probably” a risk factor. That opens up another can of worms because industry-sponsored research fails to show an association between exposure and disease. Oh well, I am not surprised there.

Lets put things in perspective. We are trapped in our web of confirmation biases. Let’s focus on better ideas (pardon my cliche) for “cross-pollination” of disciplines. Radiation Therapy is curative and is the most critical determinator of survivorship.

Can you rely on Twitter?

Of late, my engagement with Twitter has decreased as my cynicism about social media has resurfaced. I have always held the belief that Twitter is, at first, a link-sharing service. The 140 characters and URL shortening services came out of that. However, they have actively tried to increase engagement.

It doesn’t happen like that. The rates of engagement (defined by clicking on the shared links) are abysmal. It means that anyone given the user is at the mercy of algorithms. Any change in that and the discoverability falls to zero.

The Twitter timeline is unsuitable for orderly consumption. I had mentioned before, and I reiterate- it depends on how algorithm ranks your association and engagement with other users. I have tried, with varying levels of success, to participate in the “live-tweeting”, but the heterogeneous nature of discussion doesn’t add structure.

I have bet my horses on Telegram instead. It is growing, without any direct marketing. The groups remain functional by use of bots which automate policing the channel. It ensures that no one misuses the allotted privileges to speak up. I have been managing a group recently which makes it easier for disparate users to discuss issues cohesively and understandably. Sharing links, inline players and in-app browser (or instant view) is a huge plus.

If the bottom line is efficiency, then yes, Telegram wins hands down. Twitter is becoming a mass of super-added mess. I use the offline Tweet clients because the web-version has a subpar experience. Besides, it tracks using cookies and other means.

I have tried (valiantly!) to convince users to switch gears. Staying online makes it worse for identity thefts. People implicitly trust social networks, but it is a decision that is fraught with danger.

Twitter is in search of a business model that would pay up for itself. As an ad-supported service, the users are the product. Despite the real-time insights, Twitter has been dumb enough not to be able to capitalise on the generated data. Either way, despite the promises of being able to provide a medium of discovery, the real fun happens in closed groups where we chat up, in detail, about issues that are close to heart.

The new promised updates to Twitter will still languish and leave you at the mercy of nameless, faceless algorithm. Think about it.

There’s still time to change gears.

Goals of research

There has been an outpouring of dollars in basic molecular research. Many clinicians have joined in with their labs to push for “clinically relevant research”. It is evident that there would be a lot of duplication and overlap between it.

For example, look at IDH gene in the pathogenesis of gliomas. We know it carries a prognostic significance. We also know about the molecular pathogenesis. How does duplicating the research across different labs helps us or makes us any wiser?

The answer lies in the pharmaceutical business goldmine. Loath to spend on basic research in molecular pathways, the research, instead has been farmed out to a network of labs. It is easy for anyone to form a company and then sell out by being acquired. It is excellent for research ecosystem as it brings about new innovative ideas, but there are some serious issues here.

Public funded research gets outpriced for the end users who have contributed in no small measure to the same. They need to become more aware of these repercussions. Shrinking federal grants for public funded research means that there is no adequate oversight and auditing of the labs that are doing the same thing. These are potentially very high stakes, and patent awards can make individuals pretty rich.

I agree that these are generalisations and that this opinion isn’t set in stone. I have based the above assertion on my reading of the situation as well as verbal accounts.

What is urgently required is a partnership at all levels. It is to focus on one idea that has the potential to work in brain tumours. Pool in resources, under legal agreements, to work on the different aspects of the same problem. The idea above is more akin to a hub-and-spoke model of research. The goal is the identify molecular pathway and understand its implications for radiation therapy.

Let’s say, hypothetically, IDH gliomagenesis is the new pathway discovered. One team to work at a molecular level to identify potential inhibitory points, other to identify molecules that bring about this change. Another side to study the effect of radiation therapy and the pathway. Aggregated results would avoid duplication and overlap and lead to faster translational outcomes.

The problem is that they end up leaving radiation as an after-thought. It should change.