Imagine stepping into a time machine. You dial it back 50, maybe even 75 years, and materialize inside a pathology laboratory. Glance around: you might see old-fashioned, perhaps even monocular, microscopes lined up. Racks of glass Coplin jars stand ready for complex, manual staining sequences. There's the distinct, sharp tang of various chemicals from rows of reagent bottles. You notice shelves with large glass jars holding somewhat unsettling-looking preserved specimens, perhaps even older paraffin blocks still mounted on wooden bases. In a corner sits a heavy manual microtome, maybe even with the setup nearby for meticulously sharpening its reusable steel knife.
Now, fast forward. Visit a cutting-edge pathology lab today, and you’ll likely see digital slide scanners humming, molecular platforms churning out data, maybe even AI algorithms assisting with analysis. It is the image of modern medicine, automated and digitized.
Okay, picture those two labs – the mid-20th-century one and today's high-tech hub. Now, let's mentally push open the doors into the grossing room of both. Does the scene change as dramatically between eras as it did in the main lab? Let's be honest: the cutting boards, the familiar dissection tools, the inks and rulers, the reliance on scribbled notes... the core environment might feel uncannily similar. While diagnostic power outside that room has been utterly transformed, the critical starting point often resembles a remarkably well-preserved scene from the past. Why has this essential first step seemingly remained locked in time?
Innovation elsewhere, stagnation at the bench
In my previous article, I touched upon the integration of digital tools from grossing to diagnosis. But digging deeper reveals a stark contrast. Pathology as a specialty has a long history of embracing cutting-edge science when it enhances our diagnostic capabilities. While major transformative waves like immunohistochemistry (IHC) and genomic medicine are often cited as distinct revolutions reshaping modern practice, the pattern of integrating contemporary science started much earlier.
Think back to the late 19th and early 20th centuries. As pathology was establishing itself as a discipline, its pioneers weren't just cataloging morphology; they were active participants in the rapidly growing field of organic chemistry. They dove headfirst into this science, experimenting and ultimately developing the special stains – many bearing their names. These stains remain daily workhorses almost a century later. From the perspective of integrating the “new science” of the day. This was arguably pathology's first great revolution, driven from within by pathologists themselves.
Fast forward to the 1980s. The advent of IHC marked the next undisputed wave of transformation – perhaps more collaborative, forged with immunologists and biochemists. IHC fundamentally changed how we classify tumors, adding layers of protein expression data to morphology. Following closely on its heels, and further revolutionizing our field more recently, the rise of genomic medicine and molecular diagnostics has provided unprecedented insights into tumor biology and targeted therapies, adding another critical dimension beyond traditional histology and IHC.
And today? We stand amidst what many call the next major upheaval, powered by digital pathology and artificial intelligence (AI). Algorithms promise to aid in screening, grading, quantitation, and even predicting outcomes. It is an exciting, if sometimes daunting, frontier.
Yet, through all these transformations, one critical area seems perpetually left behind: the preanalytical phase, particularly the gross examination.
My own journey into this field underscored how easily we can overlook past innovations. While working on my medical thesis in 2018, applying deep learning algorithms to cervical smear interpretation, I felt I was exploring truly novel territory. Imagine my surprise discovering papers like the 1994 study on the PAPNET system, which used neural networks to analyze supposedly negative smears preceding high-grade lesion diagnoses. The idea wasn't brand new; the technology and implementation were just catching up.
And PAPNET wasn't the only early foray into digital methods. The core concept of using digital tools in pathology isn't new at all. Basic laboratory information systems emerged in the 80s and 90s to manage data. Even whole-slide imaging, the cornerstone of modern digital pathology, saw primitive precursors using magnetic tape storage back in the 1980s, around the same era early AI experiments like PAPNET were being conceived.
So, if digital concepts have been around in pathology for decades, why the persistent lag in grossing? It is not for a complete lack of trying. Looking back, I was surprised to find a 1999 paper by Cruz and Seixas from the prestigious IPATIMUP institute in Portugal. They described a remarkably forward-thinking system for its time. It wasn't complex AI, but a simple, effective use of available technology: integrating digital photos captured during gross examination into a structured report, using hypertext links for navigation, and even allowing links to subsequent microscopic images. It was a practical, early attempt at creating a comprehensive digital record starting right at the gross bench.
This reminds me of the history of the steam engine. We often credit James Watt whose work in the 18th century undeniably powered the Industrial Revolution. Yet, he wasn't working in a vacuum; the core principle wasn't entirely novel. Centuries before Watt, figures like Taqi al-Din, a key figure from the late Islamic Golden Age working in 16th-century Ottoman Constantinople, had already described early steam turbines. And going back even further, as far back as the 1st century AD, the Greco-Egyptian engineer Hero of Alexandria described a steam-powered device, the aeolipile. So, the essential concept existed long before the conditions were right for its widespread adoption to reshape the world.
Innovation often requires not just the invention, but the right ecosystem, need, and implementation strategy to truly take hold. Has grossing technology been our aeolipile – conceived but waiting for its industrial revolution?
Roadblocks and realizations
Part of the challenge lies in the unique nature of the grossing process itself. As highlighted by work like that of Véronique Hofman and colleagues, grossing occupies a peculiar temporal space in the lab workflow. Unlike many automated tests with rapid turnaround times, processing a surgical specimen can span days, involving fixation, decalcification, meticulous dissection, description, and sampling. Crucially, quality and traceability – factors vital for accurate diagnosis and downstream molecular testing – heavily rely on human skill, diligence, and robust tracking. Variables like warm and cold ischemia times, specimen handling, and accurate block labeling are critical preanalytical factors profoundly impacting diagnosis, yet often managed through manual processes and human vigilance. There have certainly been attempts to improve this – barcode systems, RFID tags, specimen tracking software interfacing with laboratory information systems – but seamless, universally adopted solutions remain elusive.
My personal "Aha!" moment regarding the potential here came about ten years ago, as a pathology resident with a keen interest in technology. I stumbled upon an article discussing the 3D printing of a human pathology specimen. It struck me as an incredible technology, but presented mainly for its educational potential. My mind raced: what if this could be integrated into the diagnostic workflow itself? Capturing the 3D structure before dissection?
Then came the video that, metaphorically speaking, blew my mind. It was Eric Glassy's talk, "Riffs on Future Path: The Fall of Paper, the Rise of Smarties and the Quest for Selfies" (available on YouTube). Watching it, I had that unsettling, almost paranoid feeling, thinking, "He's describing exactly what I've been imagining! How did he capture my thoughts?" Glassy painted a vivid picture of a digitally integrated future for pathology, starting right from the specimen's arrival.
The final piece of the puzzle clicked into place when I discovered through that same video ecosystem that an Australian pathologist, Shane Battye, had already begun actively developing concepts very similar to the ones I thought were uniquely mine, years before. It was a humbling reminder: any idea one human can conceive, another likely can too. The real challenge, the true innovation, lies not just in the conception, but in the relentless pursuit and execution required to bring that idea to fruition.
Signs of life in the digital desert of grossing
But just when it seemed the grossing room might remain perpetually marooned in its analog past, several technological oases have begun to appear on the horizon. The narrative of stagnation isn't the whole story; the digital desert is finally showing signs of life.
Perhaps the most vibrant shoots of green are emerging from computer vision. Pioneering teams are now applying AI-driven vision systems directly to the gross bench. Imagine systems assisting with specimen identification, automatically measuring specimens of all sizes, recognizing lesions or specific patterns, ensuring tissue fragments aren't missed, and preventing specimen mix-ups. These tools offer the potential for a consistent, digital second set of eyes, focused on quality and efficiency. Indeed, early results from specific automated grossing setups suggest they might process certain specimens significantly faster and perhaps more consistently than manual methods – turning concepts that felt like science fiction just years ago into emerging practical tools.
Another promising mirage-made-real involves 3D scanning and mapping. The technology is rapidly evolving beyond early demonstrations with desktop scanners. We're now seeing the potential for capturing a specimen's complex surface topography and digitally mapping crucial details like inked margins in near real-time, without cumbersome turntables. For me, as that once-curious, technophile resident, seeing this capability move from an ambitious idea to tangible reality is incredibly fulfilling. This isn't just about creating a pretty picture; it is about providing surgeons with an interactive digital model, linked within the pathology report, allowing them to rotate and visualize exactly where each diagnostic section originated on the resected organ.
Even the painstaking task of lymph node retrieval is seeing technological innovation. Techniques utilizing short-wave infrared imaging, for instance, leverage differences in how tissues interact with specific infrared wavelengths. These systems aim to make nodes stand out against surrounding fat, potentially assisting the search. This represents another front where technology is actively probing the challenges of the grossing room.
And then there are generative AI and large language models, the AI powerhouses currently transforming countless fields. Don't be surprised if your gross bench workstation soon features integrated AI assistants. In a space that has long lacked sophisticated digital support tools, these natural language helpers might just become the grossing staff's invaluable new partner.
If these developments sound faintly familiar, it might be because visionaries like Glassy have been painting this picture for years. Glassy mused about smart guidance systems, integrated digital assistants, "driverless microscopes," augmented reality, and wearable labs – ideas that seemed almost fanciful at the time. Fast forward to today, and many of those concepts are finding echoes in the tools now materializing on our benches. The computer vision systems, the 3D digital mapping, the potential AI scribes – these are, in essence, the smart laboratory assistants Glassy envisioned, finally arriving.
This wave of innovation signifies more than just acquiring cool gadgetry; it represents the long-overdue modernization of a fundamental part of our diagnostic process. The digital desert of grossing is, at last, beginning to bloom. After decades of slow percolation, we suddenly have technologies capable of irrigating this space.
The call to action for the pathology community is clear: let's actively cultivate these shoots of progress. We need to engage with these tools, evaluate them critically but with an open mind, and work collaboratively to integrate the best of them into our workflows. By welcoming this wave of smart assistance, we can finally transform our gross rooms from analog time capsules into thriving, tech-enabled hubs that enhance quality, efficiency, and ultimately, patient care. The oasis is in sight; it is time to embrace it.
Newsletters
Receive the latest pathology news, personalities, education, and career development – weekly to your inbox.
