The research team, led by radiation oncologist W. Gehin, set out to measure whether a short, structured teaching session could actually improve how accurately radiation oncology residents delineate organs at risk, and whether any improvement would show up in the numbers that matter most for treatment planning. Ten residents took part in the exploratory, single-institution study. Each of them independently contoured the organs at risk on a demanding clinical scenario: a left-sided breast cancer case requiring irradiation of the breast and the regional lymph nodes. This particular case was chosen deliberately, because it involves a dense cluster of critical structures, including the heart, the coronary arteries, the brachial plexuses, and the lung, all of which sit close to the target volume that must receive a curative radiation dose.

The study design was straightforward but rigorous in its measurement approach. Each resident contoured the case twice: once before a two-hour targeted teaching session, and again seven weeks afterward, with an eight-week interval separating the two teaching sessions. There was no control group, a limitation the authors acknowledge openly. The reference contours against which the residents were judged were created by a senior radiation oncologist following routine clinical practice at the institution, which involves reviewing and correcting the output of an AI-assisted auto-segmentation system, MVision AI version 1.2.7. A second commercial system, RayStation version 23B, was evaluated descriptively as a contextual benchmark, giving the researchers a sense of how resident performance compared with the tools now entering routine use.

To quantify geometric accuracy, the team used two well-established metrics. The Dice Similarity Coefficient, or DSC, measures the spatial overlap between two contours on a scale from zero, meaning no overlap at all, to one, meaning perfect agreement. The 95th-percentile Hausdorff Distance, or HD95, captures how far the surface of a resident’s contour strays from the reference contour at its worst local deviations, making it particularly sensitive to errors at the tips of thin or elongated structures. Together, these two metrics provide a complementary picture: DSC rewards overall volumetric agreement, while HD95 exposes localized geometric misses that a volume-based score might dilute.

The results were encouraging. In a mixed-effects statistical analysis that pooled all structures, training was associated with a significant improvement in contouring accuracy. The average DSC increased by 0.034, with a 95 percent confidence interval ranging from 0.017 to 0.052, and a p-value below 0.001. The HD95 told a parallel story: the geometric ratio of 0.788, with a confidence interval of 0.694 to 0.896, corresponds to a 21.2 percent reduction in the worst-case surface deviation after training. In practical terms, residents’ contours hugged the expert reference more tightly, and their largest local errors shrank by roughly a fifth after just two hours of focused instruction.

Perhaps the most striking findings emerged when the researchers broke the results down structure by structure. The largest improvements appeared for the brachial plexuses, the bundles of nerves that run from the spine through the neck and shoulder into the arm, with effect sizes expressed as rank-biserial correlations of 0.93 and 0.82 for the two sides. The left anterior descending coronary artery, a slender vessel whose accidental irradiation has been linked to long-term cardiac toxicity in breast cancer survivors, showed the largest improvement in HD95, with a rank-biserial correlation of 0.85. These are exactly the anatomically complex, small, and clinically consequential structures where manual contouring errors are most likely and where automated tools have historically struggled most.

However, the authors are careful to temper enthusiasm with statistical honesty. Because the study tested multiple structures simultaneously, they adjusted for multiple comparisons, and after that adjustment none of the structure-specific differences remained statistically significant. The same was true of the dosimetric analysis. To assess whether better contours translated into different dose reporting, the team applied a single, fixed volumetric-modulated arc therapy dose distribution to every contour set and recalculated the dose-volume metrics without re-optimizing the plan. This design choice is important: it isolates the effect of contour variation on the reported dose-volume values rather than implying any change in the dose actually delivered to a patient. The dosimetric differences observed were small, and none survived adjustment for multiple testing. Overall compliance with mandatory dose constraints did not change significantly either.

One of the most thought-provoking aspects of the study is its implicit benchmark against the machines. After training, the residents’ contouring accuracy was broadly comparable to that of the evaluated commercial auto-segmentation system, which notably performed less well on the small tubular structure available for comparison. This finding cuts both ways. On one hand, it suggests that a modest educational investment can bring trainees to a level of geometric performance similar to commercial software, at least on a single challenging case. On the other hand, it underscores that current AI tools are not infallible, particularly for fine, tortuous anatomy like coronary arteries and nerve plexuses, which is precisely where a trained human eye retains its value.

The authors are equally candid about the limits of their design. With no control group and with residents reassessed on the same case after training, the observed improvements should be interpreted as associations rather than as proof of a causal effect. Practice effects, familiarity with the case, or the simple passage of time could all have contributed. Moreover, the study involved only ten residents at a single institution, and the dosimetric changes, while directionally consistent with the geometric gains, were not demonstrably clinically meaningful. The team frames the observed changes as potentially relevant to plan evaluation, the process by which physicians review and approve a treatment plan, rather than as evidence of altered patient outcomes.

What the study does establish is that contouring skill is teachable, quickly, and measurably, even in an era when software promises to do the job automatically. The authors leave the deepest question deliberately open: whether the ability to critically appraise and correct AI-generated contours requires dedicated instruction of its own, or whether it develops naturally from conventional contouring training. As auto-segmentation becomes routine in clinics worldwide, that question will only grow in urgency. For now, the message from Nancy is clear and quietly reassuring. Two hours of structured teaching made residents measurably better at one of radiotherapy’s most fundamental tasks, and the structures that benefited most were the very ones where human expertise remains hardest to replace. In the race between human training and machine automation, the two, it seems, are best run together.

Subject of Research: Structured training for organ-at-risk contouring by radiation oncology residents in the era of AI auto-segmentation

Article Title: Teaching organ-at-risk delineation in the AI era: educational impact and dosimetric relevance of a structured training intervention

Article References: Gehin, W., Huger, S., Grandgirard, N., Faivre, J. C., Bruand, M., Martz, N., Charra-Brunaud, C., Salleron, J., Meyer, C., & Py, J. (2026). Teaching organ-at-risk delineation in the AI era: educational impact and dosimetric relevance of a structured training intervention. BMC Medical Education. https://doi.org/10.1186/s12909-026-10513-2

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10513-2

Keywords: radiation oncology, organ-at-risk delineation, auto-segmentation, artificial intelligence, medical education, resident training, dosimetry, Dice Similarity Coefficient, Hausdorff Distance, VMAT, brachial plexus, breast cancer radiotherapy

Stellate ganglion block, or SGB, involves injecting a local anesthetic near the stellate ganglion, a cluster of sympathetic nerve cells located in the neck. The goal is to modulate autonomic nervous system activity, and the technique has a long history in pain management and certain circulatory disorders. More recently, it has been investigated as an option for patients with stubborn insomnia who have not responded to conventional treatments. The appeal is understandable: the procedure is minimally invasive, can be performed on an outpatient basis, and some studies suggest it may improve sleep quality and overall well-being in selected patients. But as this case demonstrates, the neck is a crowded anatomical neighborhood, and needles placed near the sympathetic chain sit close to structures that the body cannot easily spare.

The patient in question had endured six months of severe insomnia before arriving at the clinic. She was sleeping roughly four hours a night, taking about ninety minutes to fall asleep, and waking four to five times nightly. Standardized questionnaires confirmed moderate-to-severe chronic insomnia: her Pittsburgh Sleep Quality Index score was 16 and her Insomnia Severity Index score was 21. She had already tried sleep hygiene education, the sedative estazolam, and melatonin with limited benefit. Her medical history included well-controlled hypertension but no cervical spine disease, trauma, or neurological disorders. Neurological examination and cervical MRI before the procedure were normal, and she showed no clinical features suggesting sleep apnea or other sleep-related movement disorders. With conservative options exhausted, her care team proceeded with a right-sided stellate ganglion block.

The procedure itself followed the modern playbook. An experienced anesthesiologist used a high-frequency linear ultrasound transducer to visualize the anatomy at the level of the sixth cervical vertebra, identifying the carotid artery, internal jugular vein, thyroid gland, and the longus colli muscle before advancing a 22-gauge needle from the side in real time. Color Doppler imaging confirmed that no vessels lay along the planned trajectory. After negative aspiration for blood and cerebrospinal fluid, the physician slowly injected 3 milliliters of 0.2 percent ropivacaine combined with 1 milliliter of compound betamethasone, a corticosteroid added to prolong the block’s effect and dampen local inflammation. Throughout needle advancement and injection, the patient reported no radiating pain, tingling, or weakness — and the injection was completed as planned.

Then, roughly one minute after the needle was withdrawn, everything changed. The patient suddenly reported severe, electric shock-like pain radiating from her right shoulder down the lateral aspect of her upper arm, followed by weakness in shoulder abduction and elbow flexion. Because the symptoms appeared only after the injection had finished, there had been no opportunity to halt the procedure in response to warning signs. Two hours later, examination revealed Medical Research Council grade 3 out of 5 strength in the right deltoid and biceps muscles, reduced pinprick sensation over the lateral upper arm in the C5 dermatome, and a diminished biceps tendon reflex. Emergency cervical MRI ruled out disc herniation, hematoma, or any other compressive lesion that could explain the sudden deficit.

Three days after the procedure, with symptoms persisting well beyond the expected duration of the local anesthetic, electromyography was performed. The results showed abundant spontaneous electrical activity — fibrillation potentials and positive sharp waves — along with markedly reduced recruitment in the deltoid and biceps, muscles predominantly supplied by the C5 nerve root. Critically, sensory nerve action potentials were preserved, a pattern that points toward a preganglionic nerve root lesion rather than an injury to a distal peripheral nerve. The authors were careful to note, however, that the clinical picture of deltoid and biceps weakness also overlaps with partial involvement of the upper trunk of the brachial plexus, and because the patient declined dedicated brachial plexus MRI and no repeat electromyography was available, definitive anatomical localization remained out of reach. The findings were therefore interpreted as favoring C5-predominant involvement rather than proving an isolated root injury.

The mechanism of injury remains uncertain, and the authors are candid about that limitation. Real-time ultrasound never provided unequivocal evidence of an intraneural or intrafascicular injection, and the stored images, being static snapshots, could not confirm that the needle had penetrated a specific neural structure. Inadvertent needle-to-nerve contact, unintended perineural or limited intraneural spread, and injection-related neural irritation all remained plausible explanations. Previous MRI research has shown that even small-volume injectate can distribute across vertebral levels from roughly C5 to T1, lending anatomical plausibility to the idea that the solution reached nerve root territory, though the authors stress this evidence supports only possibility, not proof of what happened in this patient. The role of the betamethasone additive is likewise unclear; its use was extrapolated from other regional anesthesia settings, and direct evidence for corticosteroid adjuncts in SGB for insomnia remains thin.

The recovery trajectory offers a measure of reassurance. Treatment with neurotrophic support — mecobalamin and vitamin B1 — combined with physical therapy, including ultrashort-wave therapy and medium-frequency electrical stimulation, began promptly. Pain on a 10-point visual analogue scale fell from 8 at onset to 3 at two weeks and 1 at three months. Muscle strength in the deltoid and biceps improved from grade 3 to grade 4+ out of 5 by the three-month follow-up, sensory abnormalities had largely resolved, and, somewhat remarkably, the patient’s insomnia symptoms had also improved. Still, the absence of repeat electromyography and validated functional outcome measures means the full extent of electrophysiological recovery could not be objectively characterized.

What makes this case resonate beyond a single patient is the broader context. C5 and C6 nerve root injuries have been described after cervical spine surgery and nerve root injections, but C5-predominant neurological injury following SGB performed for a sleep disorder has rarely, if ever, been documented. As SGB attracts growing interest for psychiatric and sleep-related indications — including post-traumatic stress disorder and menopausal insomnia — the population of patients exposed to the procedure is expanding well beyond the pain clinics where it was first refined. The authors argue that neurological complications may become apparent only after the block is complete, even when patients report no warning symptoms during needle placement, which means vigilance cannot end when the needle comes out. They recommend continuous needle-tip visualization, careful puncture route selection, avoidance of excessive depth, conservative injection volumes, and consideration of alternative approaches such as the anterolateral cervical technique, which may reduce nerve root exposure in selected cases.

For patients considering SGB for insomnia, the message is not that the procedure should be abandoned, but that informed consent should include an honest discussion of neurological risk, however uncommon. New radiating pain, focal weakness, sensory loss, or reflex asymmetry after any such block should trigger immediate neurological examination, appropriate imaging, and electrophysiological testing. As a single case report, the findings cannot be generalized — anatomical variation, comorbidities, and injection characteristics all influence individual risk. But the case stands as a vivid reminder that even ultrasound-guided procedures performed by experienced hands under ideal imaging conditions can produce serious complications, and that the expanding off-label use of regional anesthesia techniques demands the same rigorous safety standards, standardized training, and post-procedural monitoring applied in their original clinical homes.

Subject of Research: Suspected C5 nerve root injury temporally associated with stellate ganglion block for refractory insomnia

Article Title: Suspected C5 Nerve Root Injury Temporally Associated With Stellate Ganglion Block for Refractory Insomnia: A Case Report

Article References: Ji, X., Niu, Z., & Wang, Y. (2026). Suspected C5 Nerve Root Injury Temporally Associated With Stellate Ganglion Block for Refractory Insomnia: A Case Report. Clinical Case Reports, 14(10), Article e73603. https://doi.org/10.1002/ccr3.73603

Image Credits: AI Generated

DOI: 10.1002/ccr3.73603

Keywords: stellate ganglion block, C5 nerve root injury, refractory insomnia, ultrasound-guided injection, electromyography, regional anesthesia, neurological complications, ropivacaine, brachial plexus, patient safety, case report, autonomic nervous system