Sleep Apnea Analysis Software: OSCAR and the Alternatives

Short version

There are basically two families of CPAP software: the manufacturer’s app (myAir for ResMed, DreamMapper for Philips Respironics) and independent tools (OSCAR, a free volunteer project, and SleepHQ, a web service). The manufacturer apps are simple and require nothing but your machine’s built-in wireless connection. The independent tools show far more detail, but OSCAR needs a desktop computer and a bit of setup, while SleepHQ works in a browser and is built around sharing your data with someone else. None of them tell you what to do with the numbers — that’s still a conversation for your sleep clinician or DME provider. This article is about what each tool shows and who it fits; if you want the step-by-step for the most detailed one, see setting up OSCAR .

Why people look at their own data at all

CPAP machines record a lot more than most people realize — how many times per hour your breathing was disrupted, how much air was leaking around your mask, how long you actually wore the thing. The American Academy of Sleep Medicine describes that data as how your provider checks whether treatment is working: whether you still have disrupted breathing, whether your mask is leaking, whether you’re using the machine. Your clinic sees some of this already, usually in summary form. Looking at it yourself doesn’t replace that relationship — it just means you show up to appointments with a real question (“my AHI spiked for four nights in a row, do you know why?”) instead of just “I’m not sleeping well.”

None of the tools below diagnose anything. They display numbers your machine already recorded. What you do with those numbers — including whether a pressure setting needs to change — is a decision for the person who prescribed your therapy.

The manufacturer apps: myAir and DreamMapper

If your machine has cellular or wifi built in, it’s probably already sending data to ResMed’s myAir or Philips’ DreamMapper (or a regional equivalent). These are the path of least resistance.

What they’re good at:

• Zero setup. If your machine is connected, the app just works.
• A simple nightly score that’s easy to glance at.
• Usage streaks and basic coaching messages, which some people find motivating early on.
• Mask fit and leak feedback in plain language.

Where they fall short:

• The detail is thin. You typically get a single leak or event number for the whole night, not a timeline of when things happened.
• In my experience, myAir’s score isn’t a direct readout of your AHI — it seems to blend things like usage and mask seal into one number, though ResMed doesn’t spell out exactly how it’s calculated.
• You can’t see individual events layered against your flow rate, so if something looks off, there’s no way to investigate further inside the app.
• No card export, no annotation, no way to combine years of data into one view.

For someone who just wants a reassuring green checkmark most mornings, this is genuinely enough. For someone trying to figure out why a night went badly, it usually isn’t.

OSCAR: the deepest view, on your own desktop

OSCAR (Open Source CPAP Analysis Reporter) is free, volunteer-maintained software that reads the raw data off your machine’s SD card and turns it into detailed graphs — flow rate waveform, pressure, leak rate, every flagged event, night by night or averaged over months.

What it’s good at:

• Event-level detail. You can see an individual apnea or hypopnea on the flow rate graph itself, not just a count.
• Long-range trends. Overlaying six months of AHI or leak data next to each other is straightforward once it’s set up.
• It’s free and runs entirely on your own computer — nothing about your sleep data goes to a company’s server unless you choose to export it somewhere.
• It reads data from many ResMed and Philips Respironics machines, plus some other brands — OSCAR’s website lists which models it supports, so check yours before you start.

Where it’s more work:

• It requires a desktop or laptop (Windows, Mac, or Linux) — no phone app.
• You have to physically move the SD card from machine to computer, or use a card reader, and import it each time. It’s not automatic.
• The learning curve is real. The first time you open it, the number of charts and settings can be overwhelming.
• Interpreting what you see still takes some patience — a “flagged event” on the graph isn’t automatically something that matters clinically.

OSCAR rewards people who like data and don’t mind a bit of setup. If that’s you, the detailed walkthrough on setting up OSCAR covers getting the SD card imported and reading your first night correctly.

SleepHQ: the browser-based, shareable option

SleepHQ takes a similar level of detail to OSCAR but delivers it through a web browser instead of desktop software. You upload data from your SD card (or, for some machines, sync it more directly), and SleepHQ generates the charts online.

What it’s good at:

• Access from any device with a browser, including a phone or tablet — no dedicated desktop needed to view the charts once they’re uploaded.
• Built for sharing. If you work with a coach, a DME, or a family member who’s helping you troubleshoot, SleepHQ makes it easy to give them a link to your data instead of a folder of screenshots.
• It can combine data from multiple machines or date ranges in one account, which is handy if you’ve switched devices.
• Good-looking, readable charts that are a bit less intimidating on first look than OSCAR’s dense dashboard.

Where it’s more work or costs money:

• Free tier is limited; features like longer data history or premium reports may sit behind a paid plan, depending on current pricing.
• Because your data lives on someone else’s server rather than only your own computer, it’s worth reading SleepHQ’s privacy policy before uploading, the same as you would with any cloud health tool.
• You still need to get the data off your machine’s SD card in the first place — SleepHQ doesn’t eliminate that step, it just takes over after that point.

SleepHQ tends to fit people who want OSCAR-level detail but hate the idea of a desktop-only tool, or who specifically want to hand their data to someone else to look at.

OSCAR vs SleepHQ vs the manufacturer apps, side by side

The SD-card workflow, briefly

Both OSCAR and SleepHQ typically start the same way, whether or not your machine also has wireless reporting turned on:

1. Turn off the machine and remove the SD card (usually a small slot on the back or side — check your device’s instructions for use for the exact location).
2. Insert the card into a computer, either directly or with a USB card reader.
3. Import the card’s contents into OSCAR, or upload it through SleepHQ’s web interface.
4. Put the card back in the machine before your next session, since the machine needs it to keep recording.

It sounds fiddly the first time and becomes routine after that. The OSCAR setup guide walks through the import step in more detail.

Which metrics actually matter

Whichever software you use, a handful of numbers come up again and again:

• AHI (Apnea-Hypopnea Index) — the number of breathing disruptions per hour, the same core metric your original sleep study used. The AASM’s overview of sleep apnea explains how this index is used to describe severity.
• Leak rate — how much air is escaping around your mask seal rather than going where it should. A lot of people on CPAP describe persistent high leak as making other numbers, including AHI, less reliable, so it’s worth reading those alongside leak data rather than in isolation. Your machine’s instructions for use say what it considers an acceptable leak range.
• Usage hours — how many hours per night the machine actually ran. Medicare and many private insurers require a minimum level of use to keep covering your equipment — around 4 hours a night on most nights during an initial trial period, per the AASM . Check with your insurer or DME provider for the rule that applies to you.
• Pressure (or pressure range, if you’re on an auto-adjusting machine) — worth watching for trends, but not something to change yourself based on what you see in software; that’s squarely a conversation for whoever manages your prescription.

A single bad night rarely means much. What’s more useful is a pattern — a leak that creeps up over weeks, or an AHI that’s fine most nights but spikes after certain evenings (alcohol, sleeping position, a cold). That kind of pattern is exactly the kind of thing worth bringing to your clinician, and it’s much easier to spot with software than by memory.

What none of these tools can tell you

It’s worth being blunt about the limits here. None of these programs, including OSCAR at its most detailed, can:

• Tell you whether your current pressure setting is correct for you.
• Diagnose a new or worsening condition.
• Replace an in-lab or home sleep study if your clinician thinks one is needed.
• Reliably distinguish between an event that matters and a normal, harmless blip — that distinction often needs clinical judgment, not just a graph.

Reading your own data is useful for building a better mental model of your therapy and noticing patterns worth raising. It’s not a substitute for the person who prescribed your therapy, and it’s not a basis for changing pressure or other settings on your own.

When to talk to your doctor or DME

Bring your data — screenshots, exported reports, or a SleepHQ link — to that conversation rather than trying to interpret it alone, especially if you notice:

• A steady upward trend in AHI over several weeks rather than a single off night.
• Leak rate that’s consistently high, particularly if it lines up with mask discomfort or a switch to a new mask.
• Usage hours dropping off, which is often the first sign something about the setup — mask fit, pressure comfort, or something else — isn’t working for you.
• Any symptom that doesn’t match what the data seems to show — daytime sleepiness despite good-looking numbers, for instance, which is worth mentioning even if the software seems to say things are fine.

Your clinician may already be seeing your data through their own portal, and they’re the ones who can actually change your prescription if the data supports it. The software’s job is to help you ask a sharper question — not to give you the answer.